[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"blog-ja-quantum-portfolio-optimization-business-case":3,"blog-translated-locales-quantum-portfolio-optimization-business-case":721,"blog-all-ja":723},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"author":11,"tags":12,"readingTime":16,"body":17,"_type":715,"_id":716,"_source":717,"_file":718,"_stem":719,"_extension":720},"\u002Fblog\u002Fquantum-portfolio-optimization-business-case","blog",false,"","How to Build a Business Case for Quantum Portfolio Optimization","PortfolioQ is a small open-source tool that runs classical, penalty-method QAOA, and constraint-preserving QAOA side by side on real portfolio data, and produces the one chart every quantum business case needs.","2026-08-01","FreeQuantumComputing",[13,14,15],"Optimization","QAOA","Use Cases",9,{"type":18,"children":19,"toc":708},"root",[20,37,58,65,103,109,114,124,149,183,188,194,199,216,245,251,277,311,520,620,637,647,653,665,702],{"type":21,"tag":22,"props":23,"children":24},"element","p",{},[25,28,35],{"type":26,"value":27},"text","Writing a QAOA circuit is a solved problem. Dozens of tutorials, including ",{"type":21,"tag":29,"props":30,"children":32},"a",{"href":31},"\u002Fblog\u002Fqaoa-max-cut-tutorial",[33],{"type":26,"value":34},"our own",{"type":26,"value":36},", will get you there in an afternoon. Writing the two-page memo that convinces a risk committee to fund a pilot is a completely different skill, and almost nothing teaches it. Most \"quantum computing for finance\" content either stays purely technical (build a circuit, print a result) or purely promotional (quantum will transform portfolio optimization). Neither produces the artifact a real business case needs.",{"type":21,"tag":22,"props":38,"children":39},{},[40,48,50,56],{"type":21,"tag":29,"props":41,"children":45},{"href":42,"rel":43},"https:\u002F\u002Fgithub.com\u002FAlinDanielFerenczi\u002FPortfolioQ",[44],"nofollow",[46],{"type":26,"value":47},"PortfolioQ",{"type":26,"value":49}," is a small, open-source FastAPI project built specifically to produce that artifact. It's worth walking through, not because it's a production optimizer (its own README says explicitly that it isn't), but because its structure is a genuinely useful template for how to evaluate ",{"type":21,"tag":51,"props":52,"children":53},"em",{},[54],{"type":26,"value":55},"any",{"type":26,"value":57}," quantum optimization claim honestly, portfolio selection or otherwise.",{"type":21,"tag":59,"props":60,"children":62},"h2",{"id":61},"the-problem-it-solves",[63],{"type":26,"value":64},"The problem it solves",{"type":21,"tag":22,"props":66,"children":67},{},[68,70,76,78,85,87,93,95,101],{"type":26,"value":69},"PortfolioQ targets ",{"type":21,"tag":71,"props":72,"children":73},"strong",{},[74],{"type":26,"value":75},"cardinality-constrained portfolio selection",{"type":26,"value":77},": pick exactly K assets out of N candidates to minimize risk and maximize expected return, using the standard Markowitz mean-variance formulation: minimize ",{"type":21,"tag":79,"props":80,"children":82},"code",{"className":81},[],[83],{"type":26,"value":84},"risk_factor * wᵀΣw - μᵀw",{"type":26,"value":86}," subject to ",{"type":21,"tag":79,"props":88,"children":90},{"className":89},[],[91],{"type":26,"value":92},"sum(w) == budget",{"type":26,"value":94},", with each ",{"type":21,"tag":79,"props":96,"children":98},{"className":97},[],[99],{"type":26,"value":100},"w",{"type":26,"value":102}," binary. That constraint (exactly K assets, not \"up to K\") is what makes this a good test case: it's a real, common portfolio-construction requirement, and it's exactly the kind of constraint that's easy to get wrong in a naive QAOA implementation.",{"type":21,"tag":59,"props":104,"children":106},{"id":105},"three-solvers-run-side-by-side",[107],{"type":26,"value":108},"Three solvers, run side by side",{"type":21,"tag":22,"props":110,"children":111},{},[112],{"type":26,"value":113},"This is the part most quantum finance demos skip, and it's the part that matters for a business case: PortfolioQ doesn't run one method and report a number. It runs three, on the same data, and compares them.",{"type":21,"tag":22,"props":115,"children":116},{},[117,122],{"type":21,"tag":71,"props":118,"children":119},{},[120],{"type":26,"value":121},"Classical exact enumeration.",{"type":26,"value":123}," Brute-forces every feasible K-of-N combination directly. It's not a strawman. For portfolios in the 10-30 asset range this is fast and exact, which is precisely why it's the right baseline. Any quantum approach has to be measured against this, not against \"no answer at all.\"",{"type":21,"tag":22,"props":125,"children":126},{},[127,132,134,140,142,147],{"type":21,"tag":71,"props":128,"children":129},{},[130],{"type":26,"value":131},"Penalty-method QAOA.",{"type":26,"value":133}," The standard textbook approach: convert the constrained problem to a QUBO via ",{"type":21,"tag":79,"props":135,"children":137},{"className":136},[],[138],{"type":26,"value":139},"QuadraticProgramToQubo",{"type":26,"value":141},", encode it as an Ising Hamiltonian, run ",{"type":21,"tag":29,"props":143,"children":145},{"href":144},"\u002Fglossary\u002Fqaoa",[146],{"type":26,"value":14},{"type":26,"value":148},". It works, but it spends circuit depth and classical tuning effort enforcing the budget constraint through a penalty term, and some fraction of measured samples violates the constraint outright, silently scored badly rather than rejected. If you're presenting quantum results to people without an SDK background, this is the mode most likely to embarrass you: \"why does the quantum solution sometimes pick the wrong number of stocks?\"",{"type":21,"tag":22,"props":150,"children":151},{},[152,157,159,165,167,173,175,181],{"type":21,"tag":71,"props":153,"children":154},{},[155],{"type":26,"value":156},"Dicke-state \u002F XY-mixer QAOA.",{"type":26,"value":158}," The constraint-preserving alternative. A Dicke-state initialization (built with a small ancilla \"running count\" register that provably lands on exactly ",{"type":21,"tag":79,"props":160,"children":162},{"className":161},[],[163],{"type":26,"value":164},"|budget⟩",{"type":26,"value":166},") combined with an XY-mixer (",{"type":21,"tag":79,"props":168,"children":170},{"className":169},[],[171],{"type":26,"value":172},"RXX",{"type":26,"value":174},"\u002F",{"type":21,"tag":79,"props":176,"children":178},{"className":177},[],[179],{"type":26,"value":180},"RYY",{"type":26,"value":182}," gates) keeps every single measurement inside the feasible K-of-N subspace by construction. No penalty tuning, no invalid answers to explain away. It's a heavier circuit, but it's the version you'd want in front of a non-technical audience. Every result respects the constraint they asked for, because the ansatz makes anything else geometrically impossible.",{"type":21,"tag":22,"props":184,"children":185},{},[186],{"type":26,"value":187},"That comparison (one classical baseline, two structurally different quantum approaches) is the shape a credible business case needs. A single quantum result with no classical comparison and no discussion of why the encoding was chosen isn't evidence of anything.",{"type":21,"tag":59,"props":189,"children":191},{"id":190},"the-chart-thats-the-actual-deliverable",[192],{"type":26,"value":193},"The chart that's the actual deliverable",{"type":21,"tag":22,"props":195,"children":196},{},[197],{"type":26,"value":198},"Here's the part of PortfolioQ's own documentation worth repeating verbatim, because it's the most honest sentence you'll read about applied quantum optimization this year:",{"type":21,"tag":200,"props":201,"children":202},"blockquote",{},[203],{"type":21,"tag":22,"props":204,"children":205},{},[206,208,214],{"type":26,"value":207},"Run the same request against a real IBM backend and chart result quality vs. gate count as you increase ",{"type":21,"tag":79,"props":209,"children":211},{"className":210},[],[212],{"type":26,"value":213},"reps",{"type":26,"value":215},". This is the \"signal vs. noise crossover\" chart that's the actual deliverable for a business-case pitch.",{"type":21,"tag":22,"props":217,"children":218},{},[219,221,227,229,235,237,243],{"type":26,"value":220},"That's the whole exercise. Not \"we ran QAOA and it worked.\" Not \"quantum is X% faster.\" A chart with two axes: how close the quantum answer gets to the true optimum (",{"type":21,"tag":79,"props":222,"children":224},{"className":223},[],[225],{"type":26,"value":226},"pct_of_optimal",{"type":26,"value":228},": 100% means it matched classical exactly), against how much circuit you had to run to get there (",{"type":21,"tag":79,"props":230,"children":232},{"className":231},[],[233],{"type":26,"value":234},"circuit_depth",{"type":26,"value":236}," and ",{"type":21,"tag":79,"props":238,"children":240},{"className":239},[],[241],{"type":26,"value":242},"two_qubit_gates",{"type":26,"value":244},", both reported alongside every result). Run it on the simulator first, then on real hardware, and watch where the lines cross, or don't.",{"type":21,"tag":59,"props":246,"children":248},{"id":247},"building-the-case-step-by-step",[249],{"type":26,"value":250},"Building the case, step by step",{"type":21,"tag":22,"props":252,"children":253},{},[254,259,261,267,269,275],{"type":21,"tag":71,"props":255,"children":256},{},[257],{"type":26,"value":258},"1. Cap your problem size honestly.",{"type":26,"value":260}," PortfolioQ's own README recommends 10-30 assets for anything run on real hardware, and explains why: current ",{"type":21,"tag":29,"props":262,"children":264},{"href":263},"\u002Fglossary\u002Fnisq",[265],{"type":26,"value":266},"NISQ",{"type":26,"value":268}," devices simply don't have the qubit count or gate fidelity to go further without the noise dominating the signal. A business case that quietly assumes a 500-stock universe on today's hardware isn't a business case, it's fiction. Start with the candidate universe you'd genuinely consider: real tickers, real historical prices (the tool fetches these automatically via ",{"type":21,"tag":79,"props":270,"children":272},{"className":271},[],[273],{"type":26,"value":274},"yfinance",{"type":26,"value":276}," if you don't supply your own).",{"type":21,"tag":22,"props":278,"children":279},{},[280,285,287,293,295,301,303,309],{"type":21,"tag":71,"props":281,"children":282},{},[283],{"type":26,"value":284},"2. Establish the in-principle baseline on a simulator.",{"type":26,"value":286}," Run ",{"type":21,"tag":79,"props":288,"children":290},{"className":289},[],[291],{"type":26,"value":292},"\u002Foptimize\u002Fcompare-all",{"type":26,"value":294}," on ",{"type":21,"tag":79,"props":296,"children":298},{"className":297},[],[299],{"type":26,"value":300},"aer_simulator",{"type":26,"value":302}," before touching real hardware. Check that ",{"type":21,"tag":79,"props":304,"children":306},{"className":305},[],[307],{"type":26,"value":308},"feasible_fraction",{"type":26,"value":310}," on the XY-mixer result is exactly 1.0. That's your sanity check that the constraint-preserving construction works as designed, with no hardware noise yet in the picture.",{"type":21,"tag":312,"props":313,"children":317},"pre",{"className":314,"code":315,"language":316,"meta":7,"style":7},"language-json shiki shiki-themes github-dark","POST \u002Foptimize\u002Fcompare-all\n{\n  \"tickers\": [\"AAPL\", \"MSFT\", \"GOOGL\", \"AMZN\", \"NVDA\"],\n  \"budget\": 2,\n  \"risk_factor\": 0.5,\n  \"reps\": 2,\n  \"shots\": 1024,\n  \"backend\": \"aer_simulator\"\n}\n","json",[318],{"type":21,"tag":79,"props":319,"children":320},{"__ignoreMap":7},[321,333,342,405,429,451,472,494,512],{"type":21,"tag":322,"props":323,"children":326},"span",{"class":324,"line":325},"line",1,[327],{"type":21,"tag":322,"props":328,"children":330},{"style":329},"--shiki-default:#E1E4E8",[331],{"type":26,"value":332},"POST \u002Foptimize\u002Fcompare-all\n",{"type":21,"tag":322,"props":334,"children":336},{"class":324,"line":335},2,[337],{"type":21,"tag":322,"props":338,"children":339},{"style":329},[340],{"type":26,"value":341},"{\n",{"type":21,"tag":322,"props":343,"children":345},{"class":324,"line":344},3,[346,352,357,363,368,373,377,382,386,391,395,400],{"type":21,"tag":322,"props":347,"children":349},{"style":348},"--shiki-default:#79B8FF",[350],{"type":26,"value":351},"  \"tickers\"",{"type":21,"tag":322,"props":353,"children":354},{"style":329},[355],{"type":26,"value":356},": [",{"type":21,"tag":322,"props":358,"children":360},{"style":359},"--shiki-default:#9ECBFF",[361],{"type":26,"value":362},"\"AAPL\"",{"type":21,"tag":322,"props":364,"children":365},{"style":329},[366],{"type":26,"value":367},", ",{"type":21,"tag":322,"props":369,"children":370},{"style":359},[371],{"type":26,"value":372},"\"MSFT\"",{"type":21,"tag":322,"props":374,"children":375},{"style":329},[376],{"type":26,"value":367},{"type":21,"tag":322,"props":378,"children":379},{"style":359},[380],{"type":26,"value":381},"\"GOOGL\"",{"type":21,"tag":322,"props":383,"children":384},{"style":329},[385],{"type":26,"value":367},{"type":21,"tag":322,"props":387,"children":388},{"style":359},[389],{"type":26,"value":390},"\"AMZN\"",{"type":21,"tag":322,"props":392,"children":393},{"style":329},[394],{"type":26,"value":367},{"type":21,"tag":322,"props":396,"children":397},{"style":359},[398],{"type":26,"value":399},"\"NVDA\"",{"type":21,"tag":322,"props":401,"children":402},{"style":329},[403],{"type":26,"value":404},"],\n",{"type":21,"tag":322,"props":406,"children":408},{"class":324,"line":407},4,[409,414,419,424],{"type":21,"tag":322,"props":410,"children":411},{"style":348},[412],{"type":26,"value":413},"  \"budget\"",{"type":21,"tag":322,"props":415,"children":416},{"style":329},[417],{"type":26,"value":418},": ",{"type":21,"tag":322,"props":420,"children":421},{"style":348},[422],{"type":26,"value":423},"2",{"type":21,"tag":322,"props":425,"children":426},{"style":329},[427],{"type":26,"value":428},",\n",{"type":21,"tag":322,"props":430,"children":432},{"class":324,"line":431},5,[433,438,442,447],{"type":21,"tag":322,"props":434,"children":435},{"style":348},[436],{"type":26,"value":437},"  \"risk_factor\"",{"type":21,"tag":322,"props":439,"children":440},{"style":329},[441],{"type":26,"value":418},{"type":21,"tag":322,"props":443,"children":444},{"style":348},[445],{"type":26,"value":446},"0.5",{"type":21,"tag":322,"props":448,"children":449},{"style":329},[450],{"type":26,"value":428},{"type":21,"tag":322,"props":452,"children":454},{"class":324,"line":453},6,[455,460,464,468],{"type":21,"tag":322,"props":456,"children":457},{"style":348},[458],{"type":26,"value":459},"  \"reps\"",{"type":21,"tag":322,"props":461,"children":462},{"style":329},[463],{"type":26,"value":418},{"type":21,"tag":322,"props":465,"children":466},{"style":348},[467],{"type":26,"value":423},{"type":21,"tag":322,"props":469,"children":470},{"style":329},[471],{"type":26,"value":428},{"type":21,"tag":322,"props":473,"children":475},{"class":324,"line":474},7,[476,481,485,490],{"type":21,"tag":322,"props":477,"children":478},{"style":348},[479],{"type":26,"value":480},"  \"shots\"",{"type":21,"tag":322,"props":482,"children":483},{"style":329},[484],{"type":26,"value":418},{"type":21,"tag":322,"props":486,"children":487},{"style":348},[488],{"type":26,"value":489},"1024",{"type":21,"tag":322,"props":491,"children":492},{"style":329},[493],{"type":26,"value":428},{"type":21,"tag":322,"props":495,"children":497},{"class":324,"line":496},8,[498,503,507],{"type":21,"tag":322,"props":499,"children":500},{"style":348},[501],{"type":26,"value":502},"  \"backend\"",{"type":21,"tag":322,"props":504,"children":505},{"style":329},[506],{"type":26,"value":418},{"type":21,"tag":322,"props":508,"children":509},{"style":359},[510],{"type":26,"value":511},"\"aer_simulator\"\n",{"type":21,"tag":322,"props":513,"children":514},{"class":324,"line":16},[515],{"type":21,"tag":322,"props":516,"children":517},{"style":329},[518],{"type":26,"value":519},"}\n",{"type":21,"tag":22,"props":521,"children":522},{},[523,535,537,543,545,551,553,559,561,566,568,574,575,580,582,587,589,594,596,602,604,610,612,618],{"type":21,"tag":71,"props":524,"children":525},{},[526,528,533],{"type":26,"value":527},"3. Move to real hardware and vary ",{"type":21,"tag":79,"props":529,"children":531},{"className":530},[],[532],{"type":26,"value":213},{"type":26,"value":534},".",{"type":26,"value":536}," This is where the actual data comes from. Point ",{"type":21,"tag":79,"props":538,"children":540},{"className":539},[],[541],{"type":26,"value":542},"\"backend\"",{"type":26,"value":544}," at a real device (",{"type":21,"tag":79,"props":546,"children":548},{"className":547},[],[549],{"type":26,"value":550},"\"ibm_torino\"",{"type":26,"value":552},", for example. See our ",{"type":21,"tag":29,"props":554,"children":556},{"href":555},"\u002Fblog\u002Fibm-quantum-free-tier-2026",[557],{"type":26,"value":558},"IBM Quantum free tier guide",{"type":26,"value":560}," if you don't have hardware access set up yet) and rerun the same comparison as you increase ",{"type":21,"tag":79,"props":562,"children":564},{"className":563},[],[565],{"type":26,"value":213},{"type":26,"value":567},". ",{"type":21,"tag":79,"props":569,"children":571},{"className":570},[],[572],{"type":26,"value":573},"qaoa_xy_pct_of_optimal",{"type":26,"value":236},{"type":21,"tag":79,"props":576,"children":578},{"className":577},[],[579],{"type":26,"value":242},{"type":26,"value":581}," climbing together across a handful of ",{"type":21,"tag":79,"props":583,"children":585},{"className":584},[],[586],{"type":26,"value":213},{"type":26,"value":588}," values ",{"type":21,"tag":51,"props":590,"children":591},{},[592],{"type":26,"value":593},"is",{"type":26,"value":595}," the chart. Because each COBYLA iteration blocks on a real hardware job, use PortfolioQ's async ",{"type":21,"tag":79,"props":597,"children":599},{"className":598},[],[600],{"type":26,"value":601},"\u002Fsubmit",{"type":26,"value":603}," + ",{"type":21,"tag":79,"props":605,"children":607},{"className":606},[],[608],{"type":26,"value":609},"\u002Fjobs\u002F{id}",{"type":26,"value":611}," endpoints rather than the blocking ones. A ",{"type":21,"tag":79,"props":613,"children":615},{"className":614},[],[616],{"type":26,"value":617},"maxiter",{"type":26,"value":619}," of 100 means up to 100 hardware round-trips per optimization run, which will outlast most HTTP client timeouts on a queued device.",{"type":21,"tag":22,"props":621,"children":622},{},[623,628,630,635],{"type":21,"tag":71,"props":624,"children":625},{},[626],{"type":26,"value":627},"4. Report the noise honestly, including where it breaks the guarantee.",{"type":26,"value":629}," Expect ",{"type":21,"tag":79,"props":631,"children":633},{"className":632},[],[634],{"type":26,"value":308},{"type":26,"value":636}," to drop below 1.0 on real hardware even for the XY-mixer, despite the simulator showing a clean 1.0. That's not a bug and it's not a reason to hide the result. On real hardware, noise, not the algorithm, is what breaks the Hamming-weight preservation the Dicke-state construction guarantees in theory. That gap between simulator and hardware feasibility is itself a meaningful, quotable data point: it's a direct, measured picture of how much today's error rates cost you, expressed in a metric a risk committee reads directly.",{"type":21,"tag":22,"props":638,"children":639},{},[640,645],{"type":21,"tag":71,"props":641,"children":642},{},[643],{"type":26,"value":644},"5. Write the recommendation the data supports.",{"type":26,"value":646}," For 10-30 assets on current hardware, that recommendation is almost never \"replace the greedy heuristic in production today\". PortfolioQ's own greedy baseline exists precisely as \"what we'd run in production today,\" and it's fast, classical, and usually close to optimal at this scale. The honest, defensible business case is: here is the exact crossover point we measured, here is what has to improve (gate fidelity, qubit count, queue economics) before it moves, and here is the monitoring plan for re-running this comparison as hardware improves. That's a real deliverable. \"We're not there yet, and here's precisely how we'll know when we are\" is a stronger memo than an inflated claim that won't survive the first follow-up question.",{"type":21,"tag":59,"props":648,"children":650},{"id":649},"what-this-template-generalizes-to",[651],{"type":26,"value":652},"What this template generalizes to",{"type":21,"tag":22,"props":654,"children":655},{},[656,658,663],{"type":26,"value":657},"None of the above is specific to portfolios. The same shape (a credible classical baseline, more than one quantum encoding compared honestly, a simulator sanity check before hardware, and a quality-vs-noise chart instead of a single cherry-picked run) is exactly how you'd build a defensible case for QAOA on any other constrained optimization problem: routing, scheduling, or the Max-Cut-style problems in our ",{"type":21,"tag":29,"props":659,"children":660},{"href":31},[661],{"type":26,"value":662},"QAOA tutorial",{"type":26,"value":664},". The specific circuits change. The discipline of comparing against what you already have, and reporting where the noise wins, doesn't.",{"type":21,"tag":22,"props":666,"children":667},{},[668,670,676,678,684,686,692,694,700],{"type":26,"value":669},"If you want to see this pattern extended, PortfolioQ's ",{"type":21,"tag":29,"props":671,"children":673},{"href":42,"rel":672},[44],[674],{"type":26,"value":675},"source on GitHub",{"type":26,"value":677}," is small enough to read end to end in an afternoon: ",{"type":21,"tag":79,"props":679,"children":681},{"className":680},[],[682],{"type":26,"value":683},"portfolio.py",{"type":26,"value":685}," for the QUBO construction, ",{"type":21,"tag":79,"props":687,"children":689},{"className":688},[],[690],{"type":26,"value":691},"xy_mixer_solver.py",{"type":26,"value":693}," for the constraint-preserving ansatz, and ",{"type":21,"tag":79,"props":695,"children":697},{"className":696},[],[698],{"type":26,"value":699},"main.py",{"type":26,"value":701}," for how the comparison endpoints are wired together. It's explicitly a pilot and capability-building scaffold rather than a production system, which, for the purpose of building a business case rather than shipping a trading system, is exactly the point.",{"type":21,"tag":703,"props":704,"children":705},"style",{},[706],{"type":26,"value":707},"html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}",{"title":7,"searchDepth":335,"depth":335,"links":709},[710,711,712,713,714],{"id":61,"depth":335,"text":64},{"id":105,"depth":335,"text":108},{"id":190,"depth":335,"text":193},{"id":247,"depth":335,"text":250},{"id":649,"depth":335,"text":652},"markdown","content:blog:quantum-portfolio-optimization-business-case.md","content","blog\u002Fquantum-portfolio-optimization-business-case.md","blog\u002Fquantum-portfolio-optimization-business-case","md",[722],"en",[724,877,1028,1310,1407,1481,1616,1721,1958,2115,2238,2391,2508,2610,3146,3298,3388,3496,4422,7580,9322,9781,10889,11463,11962,14217,16076,16457,18559,19583,20602,21449,23915,24967,27514,29327,32065,34062,36164],{"_path":725,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":726,"description":727,"date":10,"author":11,"tags":728,"readingTime":474,"body":732,"_type":715,"_id":874,"_source":717,"_file":875,"_stem":876,"_extension":720},"\u002Fja\u002Fblog\u002Fquantinuum-helios-logical-qubits-2026","Quantinuum の Helios:2:1符号化による48論理量子ビットを解説","Quantinuum の最新イオントラップシステムは、わずか約96個の物理量子ビットから48個の誤り訂正済み論理量子ビットを実現したと主張しています。注目すべきは量子ビット数ではなく、その比率です。",[729,730,731],"Error Correction","Hardware","Trapped-Ion",{"type":18,"children":733,"toc":868},[734,746,757,763,775,780,792,820,825,830,835,840],{"type":21,"tag":22,"props":735,"children":736},{},[737,739,744],{"type":26,"value":738},"Quantinuum は最新のイオントラップシステム Helios を発表し、約 ",{"type":21,"tag":71,"props":740,"children":741},{},[742],{"type":26,"value":743},"2:1",{"type":26,"value":745}," の比率で符号化された48個の誤り訂正済み論理量子ビットを実現したと主張しています——約96個の物理量子ビットが48個の論理量子ビット分の仕事をしているということです。この発表に関する報道の多くは量子ビット数を主な見出しにしていました。しかし本当に注目すべき数字は、この比率のほうです。",{"type":21,"tag":22,"props":747,"children":748},{},[749,755],{"type":21,"tag":29,"props":750,"children":752},{"href":751},"\u002Fblog\u002Flogical-qubits-fault-tolerance-explained",[753],{"type":26,"value":754},"論理量子ビットと耐障害性についての関連記事",{"type":26,"value":756},"を読んでいれば、その理由はすでにお分かりでしょう。超伝導ハードウェア上の表面符号(サーフェスコード)の見積もりでは、基盤となるハードウェアが誤り閾値をどれだけ下回っているかによって、論理量子ビット1個あたり物理量子ビットが数百から数千個必要になるのが一般的です。ベンダーが報告する2:1に近い比率は、その数字の段階的な改善ではありません——設計空間のまったく別の領域に属するものであり、その数字を額面通りに受け取る前に、何がそれを可能にしているのかを理解する価値があります。",{"type":21,"tag":59,"props":758,"children":760},{"id":759},"helios-とは何か",[761],{"type":26,"value":762},"Helios とは何か",{"type":21,"tag":22,"props":764,"children":765},{},[766,768,773],{"type":26,"value":767},"Helios は Quantinuum の最新の H-Series 世代イオントラップシステムです——このラインナップはこれまで何度も量子ボリュームの記録を保持してきており、イオントラップ分野における Quantinuum の主要な競合相手である IonQ も、これを超えるべき主要なベンチマークとし続けています。イオントラップ型ハードウェアは電磁場中に個々のイオンを捕捉し、レーザーパルスで操作します。これにより、ゲート速度(超伝導量子ビットのナノ秒に対し、マイクロ秒〜ミリ秒)を犠牲にする代わりに、誤り訂正にとって極めて重要な2つの利点——より高いゲート忠実度と、",{"type":21,"tag":71,"props":769,"children":770},{},[771],{"type":26,"value":772},"全対全接続性",{"type":26,"value":774},"(チェーン内のどのイオンも、中間の量子ビットを経由せずに他のどのイオンとも量子もつれを形成できること)——を得ています。",{"type":21,"tag":59,"props":776,"children":778},{"id":777},"なぜこの比率がここで可能なのか",[779],{"type":26,"value":777},{"type":21,"tag":22,"props":781,"children":782},{},[783,785,790],{"type":26,"value":784},"Google の Willow のような超伝導チップで使われる標準的な誤り訂正符号である表面符号は、",{"type":21,"tag":71,"props":786,"children":787},{},[788],{"type":26,"value":789},"最近接",{"type":26,"value":791},"パリティチェックを基盤としています。これは平面的な2D超伝導チップが物理的にサポートできることそのものだからです。この制約こそが、そのオーバーヘッドが非常に高い大きな理由です——現実的な誤り率に対して論理量子ビット1個を保護するには、かなりの規模の2D格子状の物理量子ビットが必要になります。",{"type":21,"tag":22,"props":793,"children":794},{},[795,797,802,804,810,812,818],{"type":26,"value":796},"イオントラップにはこの制約がありません。どのイオンも他のどのイオンとも相互作用できるため、",{"type":21,"tag":71,"props":798,"children":799},{},[800],{"type":26,"value":801},"非局所的な",{"type":26,"value":803},"パリティチェックを必要とする符号——平面チップ上では配線コストが法外に高くつくタイプのもの——が直接実装可能になります。これにより、より量子ビット効率の良い符号族(大まかに言えば qLDPC 系の構成)への道が開かれ、より遅いゲートと引き換えに、物理量子ビットから論理量子ビットへのオーバーヘッドを劇的に削減できます。当サイトの",{"type":21,"tag":29,"props":805,"children":807},{"href":806},"\u002Fhardware",[808],{"type":26,"value":809},"ハードウェア概観",{"type":26,"value":811},"と",{"type":21,"tag":29,"props":813,"children":815},{"href":814},"\u002Fcompare",[816],{"type":26,"value":817},"方式比較",{"type":26,"value":819},"ではこのトレードオフをより一般的な観点から扱っていますが、Helios は現在出荷されているものの中で、これを最も明確に示す具体例です。",{"type":21,"tag":59,"props":821,"children":823},{"id":822},"この比率が教えてくれないこと",[824],{"type":26,"value":822},{"type":21,"tag":22,"props":826,"children":827},{},[828],{"type":26,"value":829},"オーバーヘッドは一つの軸にすぎず、唯一の軸ではありません。論理量子ビット1個あたりの物理量子ビット数が少ないからといって、自動的により実用的なマシンになるわけではありません。論理量子ビットについての記事で述べたように、論理量子ビットはオーバーヘッド、アイドル時の論理誤り率、論理ゲート忠実度、論理ゲート速度、ゲートセットの普遍性という複数の指標をまとめて評価すべきです。イオントラップのゲートは超伝導ゲートよりも桁違いに遅いため、1操作あたりミリ秒かかる2:1符号化の論理量子ビットは、千倍速く動作するもののコストが高い論理量子ビットより単純に「優れている」とは言えません——正しい比較は、デコヒーレンスが追いつくまでにアルゴリズムが必要とする論理演算の回数に完全に依存します。",{"type":21,"tag":22,"props":831,"children":832},{},[833],{"type":26,"value":834},"また、この業界のあらゆるハードウェアに関する主張と同様に、独立に再現されるまではベンダーが発表した数字として読むべきです。「2:1での48論理量子ビット」は、特定の運用条件下での具体的な実証を表すものであり、一般に利用可能なすぐ使える機能を意味するものではありません。",{"type":21,"tag":59,"props":836,"children":838},{"id":837},"今学んでいる人にとっての意味",[839],{"type":26,"value":837},{"type":21,"tag":22,"props":841,"children":842},{},[843,845,850,852,858,860,866],{"type":26,"value":844},"Helios クラスのハードウェアは、",{"type":21,"tag":29,"props":846,"children":847},{"href":555},[848],{"type":26,"value":849},"IBM の Open Plan",{"type":26,"value":851}," のように無料枠のキューに並んで使えるものではありません——しかし、最終的にどのハードウェアが勝者になるとしても、そこにある考え方はそのまま応用できます。Quantinuum の TKET コンパイラはオープンソースであり、どのバックエンドを対象とするかにかかわらず今日から無料で使えます。そして本稿にある概念——符号化オーバーヘッド、接続性の制約、閾値定理——は、今後どんな発表が出てきてもそれを評価するために使う、同じ道具立てです。当サイトの",{"type":21,"tag":29,"props":853,"children":855},{"href":854},"\u002Fblog\u002Funderstanding-quantum-error-correction",[856],{"type":26,"value":857},"誤り訂正の解説記事",{"type":26,"value":859},"ではその基礎的な仕組みを扱っており、",{"type":21,"tag":29,"props":861,"children":863},{"href":862},"\u002Flandscape",[864],{"type":26,"value":865},"量子業界の全体像",{"type":26,"value":867},"ではこの方式に限らず、あらゆる方式で誰が何を構築しているかを追跡しています。",{"title":7,"searchDepth":335,"depth":335,"links":869},[870,871,872,873],{"id":759,"depth":335,"text":762},{"id":777,"depth":335,"text":777},{"id":822,"depth":335,"text":822},{"id":837,"depth":335,"text":837},"content:ja:blog:quantinuum-helios-logical-qubits-2026.md","ja\u002Fblog\u002Fquantinuum-helios-logical-qubits-2026.md","ja\u002Fblog\u002Fquantinuum-helios-logical-qubits-2026",{"_path":878,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":879,"description":880,"date":10,"author":11,"tags":881,"readingTime":474,"body":884,"_type":715,"_id":1025,"_source":717,"_file":1026,"_stem":1027,"_extension":720},"\u002Fja\u002Fblog\u002Fquantum-chemistry-protein-scale-2026","量子コンピュータが本物のタンパク質をシミュレーション——その意味とは","クリーブランド・クリニックとIBMは、IBMのHeron r2チップ上で303原子のミニタンパク質の量子・古典ハイブリッドシミュレーションを実行しました。多くのVQEデモで使われるおもちゃのような分子からの本物の飛躍であり、同時に本物の留意点も伴います。",[882,883,730],"Chemistry","VQE",{"type":18,"children":885,"toc":1018},[886,899,904,909,935,940,946,951,957,984,989,994,998],{"type":21,"tag":22,"props":887,"children":888},{},[889,891,897],{"type":26,"value":890},"ネット上で見つかるほとんどの VQE チュートリアル——",{"type":21,"tag":29,"props":892,"children":894},{"href":893},"\u002Fblog\u002Fvqe-pennylane-practical-guide",[895],{"type":26,"value":896},"当サイトのもの",{"type":26,"value":898},"を含め——は、水素や水素化リチウムをシミュレーションします。原子数がわずかな分子が選ばれるのは、それが現在のハードウェアとノイズレベルで実際に扱える範囲だからです。2026年3月、クリーブランド・クリニックと IBM は、質的にまったく異なる、はるかに大きな成果を発表しました。IBM の Heron r2 プロセッサを用いて、303原子のミニタンパク質 Trp-cage の電子構造をモデル化する量子・古典ハイブリッドワークフローです。",{"type":21,"tag":22,"props":900,"children":901},{},[902],{"type":26,"value":903},"これは規模において本物の飛躍であり、それがなぜ難しいのか、そして見出しになった原子数が何を証明していて何を証明していないのかを理解する価値があります。",{"type":21,"tag":59,"props":905,"children":907},{"id":906},"なぜ分子の大きさが問題の核心なのか",[908],{"type":26,"value":906},{"type":21,"tag":22,"props":910,"children":911},{},[912,918,920,925,927,933],{"type":21,"tag":29,"props":913,"children":915},{"href":914},"\u002Fresearch\u002Ffeynman-simulating-physics-1982",[916],{"type":26,"value":917},"ファインマンが1982年に量子計算を提唱した際の根拠",{"type":26,"value":919},"は、まさにこの点にありました——量子系を古典コンピュータ上で厳密にシミュレーションするコストは、システムサイズに対して指数関数的に増大します。なぜなら電子の波動関数は、関わる軌道数に応じて次元が爆発的に増えるヒルベルト空間の中に存在するからです。化学者は通常これを近似によって回避しますが、小さな分子を超えるものについて厳密な答えを得ることは、古典計算では手が届きません。これこそが ",{"type":21,"tag":29,"props":921,"children":923},{"href":922},"\u002Fglossary\u002Fvqe",[924],{"type":26,"value":883},{"type":26,"value":926}," の存在意義そのものです——VQE を導入した",{"type":21,"tag":29,"props":928,"children":930},{"href":929},"\u002Fresearch\u002Fperuzzo-vqe-2014",[931],{"type":26,"value":932},"Peruzzo の2014年の論文",{"type":26,"value":934},"が量子・古典ハイブリッドループを用いたのは、まさに近い将来のハードウェアが計算全体を単独ではこなせないからでした。",{"type":21,"tag":22,"props":936,"children":937},{},[938],{"type":26,"value":939},"これはまた、「原子はいくつか?」が量子化学のデモンストレーションにとって本当に意味のある指標である理由でもあります——ランダム回路サンプリングのベンチマークなどとは違う点です。2原子の分子から303原子の分子へという飛躍は、同じ結果を誇張したマーケティングではありません——取り組んでいる問題の規模における、本物の飛躍なのです。",{"type":21,"tag":59,"props":941,"children":943},{"id":942},"trp-cage-とは何かなぜ選ばれたのか",[944],{"type":26,"value":945},"Trp-cage とは何か、なぜ選ばれたのか",{"type":21,"tag":22,"props":947,"children":948},{},[949],{"type":26,"value":950},"Trp-cage はプレスリリースのために適当に選ばれた大きな分子ではありません——それは、扱いやすいほど小さく、かつ本物のタンパク質のように振る舞うほど現実的であるという理由から、長年にわたり分子動力学やタンパク質フォールディング研究の標準的なベンチマークとして使われてきた、よく研究された20残基のミニタンパク質です。新奇なシステムではなく確立された参照システムを使うことで、数十年にわたる既存の古典シミュレーションデータと結果を照合しやすくなります。",{"type":21,"tag":59,"props":952,"children":954},{"id":953},"ここでのハイブリッドワークフローがほぼ確実に意味すること",[955],{"type":26,"value":956},"ここでの「ハイブリッドワークフロー」がほぼ確実に意味すること",{"type":21,"tag":22,"props":958,"children":959},{},[960,962,967,969,974,976,982],{"type":26,"value":961},"原子数を注意深く読んでください。303原子系の",{"type":21,"tag":51,"props":963,"children":964},{},[965],{"type":26,"value":966},"完全な",{"type":26,"value":968},"電子構造は、現在のどの QPU が直接扱える範囲もはるかに超えており、303個すべての原子の電子が実際に量子プロセッサ上で表現された可能性は非常に低いといえます。現実的な描像は——IBM が同種の大規模分子研究について発表してきた手法と一致するもので——ハイブリッド埋め込み方式です。注目する性質にとって最も重要な電子と軌道からなる、慎重に選ばれたアクティブスペースが量子ハードウェア上で処理され、残りのシステムは古典的に処理されて結果が統合されます。これは成果を貶めるものではありません——このサイズ範囲における「量子コンピュータが本物の分子をシミュレーションした」という主張は、現状すべてこの方式で成り立っており、まさに",{"type":21,"tag":29,"props":970,"children":971},{"href":263},[972],{"type":26,"value":973},"NISQ 時代",{"type":26,"value":975},"においてこの分野が落ち着いた",{"type":21,"tag":29,"props":977,"children":979},{"href":978},"\u002Fglossary\u002Fhybrid-algorithm",[980],{"type":26,"value":981},"ハイブリッドアルゴリズム",{"type":26,"value":983},"のパターンそのものです。誠実に言えば、見出しの数字が表しているのは研究対象となったシステムであり、QPU がすべての原子に等しく仕事をしたという証明ではありません。",{"type":21,"tag":59,"props":985,"children":987},{"id":986},"それでも意義あるマイルストーンである理由",[988],{"type":26,"value":986},{"type":21,"tag":22,"props":990,"children":991},{},[992],{"type":26,"value":993},"留意点はあるにせよ、これらのハイブリッドパイプラインが扱える範囲における本物の進歩です。これまでの「量子コンピュータが化学をする」というデモの大半は、最小基底系で原子数がわずかな分子にとどまっていました。同じ種類の手法を、生物学的に意味があり、十分に特性評価された303原子系にまでスケールさせ、しかも量子部分を古典的な代替ではなく実際のハードウェア上で実行したことは、パイプライン自体がスケールすることを示しており、それこそが各社が競って解決しようとしている本当のボトルネックです。これは2026年の他の化学マイルストーン——たとえば Qubit Pharmaceuticals と Pasqal によるタンパク質ポケット内での水分子配置の量子アルゴリズムによるデモンストレーション(教科書的な課題ではなく、分子生物学レベルの課題)——と同じ方向に位置づけられます。",{"type":21,"tag":59,"props":995,"children":996},{"id":837},[997],{"type":26,"value":837},{"type":21,"tag":22,"props":999,"children":1000},{},[1001,1003,1009,1011,1016],{"type":26,"value":1002},"ここで使われている技術を学ぶのに、Heron r2 へのアクセスも、実機の QPU すら必要ありません。同じ VQE の構成要素(アンザッツの構築、アクティブスペースの選択、古典的な最適化ループ)は、",{"type":21,"tag":29,"props":1004,"children":1006},{"href":1005},"\u002Fsimulators",[1007],{"type":26,"value":1008},"無料のローカルシミュレータ",{"type":26,"value":1010},"で H₂ をフィッティングする場合でも、303原子のタンパク質について読む場合でも変わりません。小分子バージョンを自分で実行してみたい場合は、当サイトの ",{"type":21,"tag":29,"props":1012,"children":1013},{"href":893},[1014],{"type":26,"value":1015},"PennyLane VQE ガイド",{"type":26,"value":1017},"から始めてみてください。そして、こうした報道に出てくる見出しの原子数は、他のあらゆるハードウェアの主張と同じように扱いましょう——それは本物の成果であり、正確に理解する価値があり、そしてまだ話の全体ではないのです。",{"title":7,"searchDepth":335,"depth":335,"links":1019},[1020,1021,1022,1023,1024],{"id":906,"depth":335,"text":906},{"id":942,"depth":335,"text":945},{"id":953,"depth":335,"text":956},{"id":986,"depth":335,"text":986},{"id":837,"depth":335,"text":837},"content:ja:blog:quantum-chemistry-protein-scale-2026.md","ja\u002Fblog\u002Fquantum-chemistry-protein-scale-2026.md","ja\u002Fblog\u002Fquantum-chemistry-protein-scale-2026",{"_path":1029,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1030,"description":1031,"date":10,"author":11,"tags":1032,"readingTime":1034,"body":1035,"_type":715,"_id":1307,"_source":717,"_file":1308,"_stem":1309,"_extension":720},"\u002Fja\u002Fblog\u002Ftop-quantum-computing-companies-2026","2026年7月時点の量子コンピューティング企業ランキングと解説","2026年7月時点で本当に重要な量子コンピューティング企業を、プレスリリースの熱気ではなく実際の売上、実際のハードウェアの節目、実際の手元資金で評価したランキング。IonQの数字はその順位に見合っている。Rigettiの数字はまだ競合他社に及んでいない。",[1033,730],"Industry",12,{"type":18,"children":1036,"toc":1293},[1037,1042,1054,1060,1080,1086,1091,1097,1102,1123,1128,1134,1147,1153,1158,1164,1169,1174,1180,1185,1191,1196,1202,1207,1213,1218,1224,1262,1267],{"type":21,"tag":22,"props":1038,"children":1039},{},[1040],{"type":26,"value":1041},"「量子コンピューティング企業トップ」を謳うリストの多くは、見出しの量子ビット数か、直近のプレスリリースの新しさで順位を決めている。どちらもほとんど意味を持たない。フィデリティの数値が伴わない量子ビット数は、そのマシンが実際に役立つ仕事をしているかどうかを何も語らないし、プレスリリースはマーケティング予算を測るものであって、ハードウェアのロードマップを測るものではない。",{"type":21,"tag":22,"props":1043,"children":1044},{},[1045,1047,1052],{"type":26,"value":1046},"このランキングでは代わりに三つの基準を使う。実証された技術的進展（フィデリティ、論理量子ビット、見出しの主張だけではなく独立に再現可能な結果）、商業的な牽引力（売上、契約、受注残）、そして財務的な耐久力（資金消費率に対する手元資金の状況）だ。ある数字が企業自身の主張だけを根拠にしている場合は、その旨を明示している。当サイトの",{"type":21,"tag":29,"props":1048,"children":1049},{"href":862},[1050],{"type":26,"value":1051},"業界一覧ページ",{"type":26,"value":1053},"には各カテゴリーで60社以上を収録している。この記事はその対極にある。誰が実際に先行していて、声の大きい名前の一部がなぜそうではないのかを短く、独自の視点でまとめたリストだ。",{"type":21,"tag":59,"props":1055,"children":1057},{"id":1056},"_1-ibm-quantum",[1058],{"type":26,"value":1059},"1. IBM Quantum",{"type":21,"tag":22,"props":1061,"children":1062},{},[1063,1065,1071,1073,1078],{"type":26,"value":1064},"IBMは業界で最も幅広いプラットフォームであり続けており、2026年7月は記録に残る中でも特に力強い月だった。三つの異なるパートナーと三つの異なる検証手法による、三件の別々の量子アドバンテージ成果が同日に発表された。",{"type":21,"tag":29,"props":1066,"children":1068},{"href":1067},"\u002Fblog\u002Fibm-quantum-advantage-triple-announcement-2026",[1069],{"type":26,"value":1070},"この発表の全容はこちらで取り上げている",{"type":26,"value":1072},"。Nighthawkプロセッサ（120量子ビット、218の可変結合器）はすでに稼働しており、IBMの公開ロードマップは一連の中間システムを経て2029年までのフォールトトレランス実現を目指し、表面符号に比べて誤り訂正のオーバーヘッドを約90%削減するqLDPC系の符号への移行を進めている。IBMのOpen Planも、誰もが無料で実機に回路を実行できる最も簡単な方法であり続けており、これが次世代の量子開発者の学び方の中心にIBMを位置づけている。自分で試してみたいなら",{"type":21,"tag":29,"props":1074,"children":1075},{"href":555},[1076],{"type":26,"value":1077},"IBM Quantum無料プランのガイド",{"type":26,"value":1079},"を参照してほしい。IBMの弱点はその幅広さの裏返しでもある。これだけ多くのことが同時に起きていると、一つ一つの成果が受ける検証は、より小規模で焦点を絞った競合が受けるものより手薄になる。",{"type":21,"tag":59,"props":1081,"children":1083},{"id":1082},"_2-google-quantum-ai",[1084],{"type":26,"value":1085},"2. Google Quantum AI",{"type":21,"tag":22,"props":1087,"children":1088},{},[1089],{"type":26,"value":1090},"GoogleのWillowプロセッサは2025年末にQuantum Echoesの成果を出した（Nature誌に掲載）。論文はこの計算を、利用可能な最良の古典的手法よりも約1万3000倍高速だと見積もっている。その1年前のしきい値以下の誤り訂正実証は、この分野で完全に独立した検証を受けた数少ない節目の一つであり続けている。Googleは2026年3月、超伝導方式の研究に加えて中性原子の研究にも着手した。トランズモン量子ビットが最終的な答えではない可能性への備えだ。Googleは競合他社より発表頻度が少なく、発表と発表の間に語る情報も少ない。これは擁護できる研究戦略だが、月単位で進捗を追いにくくもしている。",{"type":21,"tag":59,"props":1092,"children":1094},{"id":1093},"_3-ionq",[1095],{"type":26,"value":1096},"3. IonQ",{"type":21,"tag":22,"props":1098,"children":1099},{},[1100],{"type":26,"value":1101},"純粋な量子ハードウェア企業の中で、IonQの数字は業界で最も強い。これは宣伝文句ではなく事実の記述だ。2026年第1四半期の売上高は6470万ドルに達し、前年同期比755%増となった。通期のガイダンスは2億6000万〜2億7000万ドルに引き上げられ、残存履行義務は4億7000万ドル、前年比554%増だった。これは希望的観測で運営している企業の姿ではない。政府部門での牽引力もそれを裏付けている。ミサイル防衛局との契約枠は上限1510億ドル（上限であって支出の保証ではないが、それでも異例の規模だ）、2026年4月には新規のDARPA契約、韓国のKISTI研究所とのソブリン量子HPC提携もある。",{"type":21,"tag":22,"props":1103,"children":1104},{},[1105,1107,1113,1115,1121],{"type":26,"value":1106},"ハードウェア面では、IonQのトラップイオン方式はOxford Ionics買収由来の「スムーズゲート」技術の恩恵を受けている。同社によれば、この技術は完全な基底状態冷却を必要とせずに2量子ビットゲートのフィデリティを99.99%超に引き上げるという。独立検証で裏付けられれば、本物の工学的近道になる。IonQはまた2025年12月、量子LDPC符号向けのデコーダー成果（Beam Search）を発表し、標準的なBP-OSDに対して論理エラー率を5.6〜17倍削減し、通常のCPUコア上でミリ秒未満のデコードを実現すると主張している。もしIonQの想定通りにスケールするなら、競合が追求しているFPGAやASICのルートよりもかなり安価なリアルタイムデコードへの道になる。この話題は",{"type":21,"tag":29,"props":1108,"children":1110},{"href":1109},"\u002Fblog\u002Fquantum-error-decoding-bottleneck",[1111],{"type":26,"value":1112},"リアルタイムデコーディングのボトルネック",{"type":26,"value":1114},"でさらに掘り下げている。そして",{"type":21,"tag":29,"props":1116,"children":1118},{"href":1117},"\u002Fblog\u002Fionq-skywater-acquisition-explained",[1119],{"type":26,"value":1120},"詳しく検証した",{"type":26,"value":1122},"SkyWater Technology買収は、IonQに量子ビット周辺の電極、フォトニクス、制御用ハードウェアについて完全に国内で完結し防衛用途の認可を受けたサプライチェーンをもたらす。量子ビットの物理特性とは関係なく、政府調達に対する本物の強みだ。",{"type":21,"tag":22,"props":1124,"children":1125},{},[1126],{"type":26,"value":1127},"これらはいずれも、Googleのしきい値以下成果ほど独立に検証されてはいない。IonQのフィデリティ、デコーダー、ゲート速度に関する主張はIonQ自身の数字だ。しかし商業面と政府部門での牽引力はマーケティングの主張とは異なり、外部から報告され監査可能なものであり、これほど若いハードウェア企業が研究資金だけでなくこの規模の実際の売上成長を示すのは珍しい。この組み合わせ、実証された商業的な引き合いと筋の通ったハードウェアロードマップこそが、IonQが同等かそれ以上のラボ成果を持ちながら商業的な実績がはるかに薄い競合他社より上位にランクされる理由だ。",{"type":21,"tag":59,"props":1129,"children":1131},{"id":1130},"_4-quantinuum",[1132],{"type":26,"value":1133},"4. Quantinuum",{"type":21,"tag":22,"props":1135,"children":1136},{},[1137,1139,1145],{"type":26,"value":1138},"QuantinuumのHeliosシステムは、約96個の物理量子ビットから48個の誤り訂正済み論理量子ビットを実現したと主張しており、2対1という比率だ。これが確認されれば重要な意味を持つ。超伝導ハードウェア上の表面符号の見積もりは通常、論理量子ビット1個あたり物理量子ビットが数百から数千個に及ぶからだ。",{"type":21,"tag":29,"props":1140,"children":1142},{"href":1141},"\u002Fblog\u002Fquantinuum-helios-logical-qubits-2026",[1143],{"type":26,"value":1144},"この比率がそもそもなぜ可能なのかはこちらで解説している",{"type":26,"value":1146},"。トラップイオン方式の全対全接続性により、平面型の超伝導チップでは配線できない非局所的なパリティチェックが可能になる。Quantinuumは2026年に156億ドルの評価額で株式公開し、それは数か月前の直近の非公開ラウンドを50%以上上回る水準だった。ハネウェルは議決権の約48%を保持している。このIPOの反応は、市場が技術的な話を信じていることを示している。IonQのフィデリティに関する主張と同様、この2対1という数字もベンダー自身の報告であり、まだ独立に再現されていない。",{"type":21,"tag":59,"props":1148,"children":1150},{"id":1149},"_5-d-wave",[1151],{"type":26,"value":1152},"5. D-Wave",{"type":21,"tag":22,"props":1154,"children":1155},{},[1156],{"type":26,"value":1157},"D-Waveは別のカテゴリーに属する。ゲート方式ではなくアニーリング方式であり、そのために過小評価されやすい。2026年第1四半期の受注は前年比でほぼ20倍に急増し3340万ドルに達したが、計上済みの売上高は比較的少額の290万ドルにとどまり、受注残は本物だが計上までの遅れが長いことを示している。5億8800万ドルの手元資金がその間を待つ余裕を与えている。フロリダ・アトランティック大学やフォーチュン100企業との量子クラウドアクセス契約は、アニーリングに自然に当てはまる最適化問題というニッチに、依然として活発な商業需要があることを示している。",{"type":21,"tag":59,"props":1159,"children":1161},{"id":1160},"_6-rigetti-computing",[1162],{"type":26,"value":1163},"6. Rigetti Computing",{"type":21,"tag":22,"props":1165,"children":1166},{},[1167],{"type":26,"value":1168},"Rigettiは、このリストの中で語られている話と実際の数字との差が最も大きい企業だ。108量子ビットの超伝導システムCepheus-1は2026年4月に一般提供に達したが、当初の2025年第4四半期という目標からずれ込んでおり、可変結合器のフィデリティ問題が原因だった。出荷後に報告された2量子ビットゲートのフィデリティは99〜99.1%程度で、Rigetti自身の目標である99.5%を下回り、IonQとQuantinuumがそれぞれのプラットフォームについて主張する数字を大きく下回る。2026年第1四半期の売上高は440万ドルで、同じ四半期のIonQの6470万ドルのごく一部に過ぎない。公開情報ではDARPAプログラムの節目での後退も報じられており、現時点でロードマップの大半を自己資金でまかなっている企業にとって、さらなる圧力となっている。",{"type":21,"tag":22,"props":1170,"children":1171},{},[1172],{"type":26,"value":1173},"真に評価できる強みは一つ、財務基盤だ。約5億6900万ドルの現金、無借金、四半期あたり約2600万ドルの資金消費率で、これは本物の活動余地を買っている。これは率直に評価する価値がある。強固な手元資金は慰めではなく、本物の資産だからだ。しかし現金が買うのは時間であって競争上の地位ではない。そして量子コンピュータが実際に役立つ仕事をするかどうかを本当に予測する指標、フィデリティ、論理量子ビットの進展、独立した商業的牽引力において、RigettiはIBM、Google、IonQ、Quantinuumと並んでいるのではなく、その後ろを走っている。1000量子ビット超を目指す同社自身の公開ロードマップは、数字を見る限り、並走する道というより数年がかりの追い上げのように読める。",{"type":21,"tag":59,"props":1175,"children":1177},{"id":1176},"_7-psiquantum",[1178],{"type":26,"value":1179},"7. PsiQuantum",{"type":21,"tag":22,"props":1181,"children":1182},{},[1183],{"type":26,"value":1184},"フォトニック方式でフォールトトレランスを最優先するPsiQuantumは、2025年9月にシリーズEで10億ドルを調達し、オーストラリア連邦政府とクイーンズランド州政府から約9億4000万オーストラリアドル（約6億2000万米ドル）の資金コミットメントを取り付けた。ブリスベンの施設の建設は2026年6月に着工したが、納入予定は当初の2027年から2029年に後退している。PsiQuantumは商用システムを一度も出荷したことがなく、フォールトトレラントなマシンが大規模に構築された初回で動作することにすべてを賭けている。このリストにあるゲート方式の競合他社とは根本的に異なるリスクプロファイルであり、それらはいずれも今日、ノイズはあるものの実際に動くハードウェアを現場に持っている。",{"type":21,"tag":59,"props":1186,"children":1188},{"id":1187},"_8-xanadu",[1189],{"type":26,"value":1190},"8. Xanadu",{"type":21,"tag":22,"props":1192,"children":1193},{},[1194],{"type":26,"value":1195},"Xanaduは2026年3月にナスダックとトロント証券取引所に上場し、3億200万ドルを調達した。そのモジュール型フォトニックシステムAuroraはリアルタイム誤り訂正を実現したと主張しており、独立に検証されれば意味のある技術的な一歩となる。同社のPennyLaneフレームワークは、どのハードウェアを対象にするかにかかわらず、量子機械学習分野で最も広く使われているオープンソースツールの一つであり続けており、自社チップが勝つかどうかに依存しないエコシステム上の足場をXanaduに与えている。",{"type":21,"tag":59,"props":1197,"children":1199},{"id":1198},"_9-pasqal",[1200],{"type":26,"value":1201},"9. Pasqal",{"type":21,"tag":22,"props":1203,"children":1204},{},[1205],{"type":26,"value":1206},"中性原子方式のPasqalは3億4000万ユーロを調達し、2026年下半期に完了予定のSPAC合併で20億ドルの評価額を付けた。公開ロードマップでは2026年に250量子ビットでのアドバンテージ実証を目指し、VelaおよびCentaurusシステムで1万量子ビット超への道筋を描いている。中性原子はトラップイオンや超伝導量子ビットに比べて商業化の段階としてはまだ早いが、Pasqalの資金力と（特に欧州における）政府との関係は、その規模に到達するための本物の持久力を与えている。",{"type":21,"tag":59,"props":1208,"children":1210},{"id":1209},"_10-quera-computing",[1211],{"type":26,"value":1212},"10. QuEra Computing",{"type":21,"tag":22,"props":1214,"children":1215},{},[1216],{"type":26,"value":1217},"QuEraは2026年1月のNature誌の論文で、448個の物理量子ビット上に96個の論理量子ビットを符号化したと報告した。この比率がGoogleのしきい値以下成果と同様の検証に耐えれば、QuEraの論理量子ビット数はQuantinuumのHelios成果を上回ることになる。Googleやソフトバンクなどからの累計5億700万ドル超の資金調達は、中性原子方式をさらなる規模へと押し進める原資となっている。",{"type":21,"tag":59,"props":1219,"children":1221},{"id":1220},"その他の企業簡単に",[1222],{"type":26,"value":1223},"その他の企業、簡単に",{"type":21,"tag":22,"props":1225,"children":1226},{},[1227,1232,1234,1239,1241,1246,1248,1253,1255,1260],{"type":21,"tag":71,"props":1228,"children":1229},{},[1230],{"type":26,"value":1231},"Atom Computing",{"type":26,"value":1233},"は、マイクロソフトのQuNorthグループと誤り訂正で協業しながら、2026年末までに1200以上の物理量子ビット上で50個の論理量子ビットを目指している。",{"type":21,"tag":71,"props":1235,"children":1236},{},[1237],{"type":26,"value":1238},"IQM",{"type":26,"value":1240},"はブラックロックから5000万ユーロを調達し、18億ドルのSPAC評価額を達成、EU初の純粋な量子ハードウェア上場企業として自らを位置づけており、すでに13社の顧客に21台のシステムを販売済みだ。",{"type":21,"tag":71,"props":1242,"children":1243},{},[1244],{"type":26,"value":1245},"Infleqtion",{"type":26,"value":1247},"は2026年2月にニューヨーク証券取引所に上場し、1600個の物理量子ビットとフィデリティ99.73%を報告している。",{"type":21,"tag":71,"props":1249,"children":1250},{},[1251],{"type":26,"value":1252},"Microsoft",{"type":26,"value":1254},"は2026年6月にトポロジカル量子ビットのMajorana 2アップデートを出荷し、量子ビットの寿命が1000倍向上したと主張しているが、トポロジカル量子ビットに関する主張が歴史的に物議を醸してきたことを踏まえ、この結果は今なお物理学界からの本物の検証にさらされている。",{"type":21,"tag":71,"props":1256,"children":1257},{},[1258],{"type":26,"value":1259},"NVIDIA",{"type":26,"value":1261},"は量子ビットを一切製造していないが、NVQLinkインフラを、Rigetti、SEEQC、Quantinuum、IQMのQPUと古典的なスーパーコンピューティングをつなぐ接続層として位置づけており、どのハードウェア方式が勝っても自社がその配線部分を握れるという賭けに出ている。",{"type":21,"tag":59,"props":1263,"children":1265},{"id":1264},"このリストの使い方",[1266],{"type":26,"value":1264},{"type":21,"tag":22,"props":1268,"children":1269},{},[1270,1272,1278,1280,1285,1286,1291],{"type":26,"value":1271},"こうしたランキングはすぐに古くなる。ここに載っている数字はどれも特定の四半期や発表に基づくものであり、1年もすれば古びて見えるだろう。それほど早く古びないのは方法論のほうだ。売上高をガイダンスと照らし合わせる、フィデリティを企業自身が掲げた目標と照らし合わせる、「論理量子ビット」や「アドバンテージ」の主張が独立に再現されたのか、それとも依然として一つのベンダーの言葉だけに乗っているのかを確認する。次に読む発表にも同じ方法を当てはめてほしい。当サイトの",{"type":21,"tag":29,"props":1273,"children":1275},{"href":1274},"\u002Fblog\u002Fquantum-benchmarking-eo-14413",[1276],{"type":26,"value":1277},"量子ベンチマークに関する記事",{"type":26,"value":1279},"では、この分野になぜいまだにベンダーから独立した信頼できるハードウェア比較の手段が存在しないのかをさらに掘り下げている。実際にどのプラットフォームの上に構築するかを決めたいなら、ただ眺めるだけの対象を選ぶのではなく、当サイトの",{"type":21,"tag":29,"props":1281,"children":1282},{"href":806},[1283],{"type":26,"value":1284},"ハードウェア概要",{"type":26,"value":811},{"type":21,"tag":29,"props":1287,"children":1288},{"href":814},[1289],{"type":26,"value":1290},"SDK比較",{"type":26,"value":1292},"が次に見るべき場所だ。",{"title":7,"searchDepth":335,"depth":335,"links":1294},[1295,1296,1297,1298,1299,1300,1301,1302,1303,1304,1305,1306],{"id":1056,"depth":335,"text":1059},{"id":1082,"depth":335,"text":1085},{"id":1093,"depth":335,"text":1096},{"id":1130,"depth":335,"text":1133},{"id":1149,"depth":335,"text":1152},{"id":1160,"depth":335,"text":1163},{"id":1176,"depth":335,"text":1179},{"id":1187,"depth":335,"text":1190},{"id":1198,"depth":335,"text":1201},{"id":1209,"depth":335,"text":1212},{"id":1220,"depth":335,"text":1223},{"id":1264,"depth":335,"text":1264},"content:ja:blog:top-quantum-computing-companies-2026.md","ja\u002Fblog\u002Ftop-quantum-computing-companies-2026.md","ja\u002Fblog\u002Ftop-quantum-computing-companies-2026",{"_path":1311,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1312,"description":1313,"date":10,"author":11,"tags":1314,"readingTime":474,"body":1315,"_type":715,"_id":1404,"_source":717,"_file":1405,"_stem":1406,"_extension":720},"\u002Fblog\u002Fadapt-gqe-ai-quantum-circuit-design","AI Now Designs Quantum Chemistry Circuits Faster Than the Standard Method","Quantinuum, NVIDIA, and Pfizer built ADAPT-GQE, a transformer and reinforcement-learning system that generates VQE ansatz circuits an order of magnitude faster than ADAPT-VQE, demonstrated on a real drug molecule and run on Quantinuum's Helios hardware.",[882,883,730],{"type":18,"children":1316,"toc":1398},[1317,1328,1334,1339,1344,1350,1361,1367,1380,1386],{"type":21,"tag":22,"props":1318,"children":1319},{},[1320,1322,1326],{"type":26,"value":1321},"Building a good ",{"type":21,"tag":29,"props":1323,"children":1324},{"href":922},[1325],{"type":26,"value":883},{"type":26,"value":1327}," circuit for a specific molecule is itself a hard, expensive search problem, before you ever run the circuit. Researchers from Quantinuum, NVIDIA, and Pfizer published a result on July 30, 2026 that automates that search: a system called ADAPT-GQE, which learns to generate the ansatz circuit directly instead of building it one operator at a time.",{"type":21,"tag":59,"props":1329,"children":1331},{"id":1330},"the-problem-adapt-gqe-targets",[1332],{"type":26,"value":1333},"The problem ADAPT-GQE targets",{"type":21,"tag":22,"props":1335,"children":1336},{},[1337],{"type":26,"value":1338},"The standard method, ADAPT-VQE, builds a molecule-specific ansatz by testing a pool of candidate operators at every step, picking the one that improves the energy estimate most, and repeating. That greedy search works, but it gets expensive fast as molecules grow, since every step re-evaluates the whole operator pool from scratch.",{"type":21,"tag":22,"props":1340,"children":1341},{},[1342],{"type":26,"value":1343},"ADAPT-GQE replaces that per-molecule greedy search with a trained model: a transformer combined with reinforcement learning, trained to generate the state-preparation circuit for a molecule's electronic ground state directly. Once trained, generating a new circuit is a forward pass through a model instead of a fresh greedy search.",{"type":21,"tag":59,"props":1345,"children":1347},{"id":1346},"what-the-results-show",[1348],{"type":26,"value":1349},"What the results show",{"type":21,"tag":22,"props":1351,"children":1352},{},[1353,1355,1360],{"type":26,"value":1354},"The team reports a 234 times speed-up in generating training data for complex molecules, and roughly one order of magnitude reduction in circuit generation time compared to ADAPT-VQE, while matching or improving accuracy in preparing molecular ground states on their benchmarks. They demonstrated the approach on imipramine, an approved tricyclic antidepressant used here as a realistic pharmaceutical test molecule, and ran the AI-generated circuits on Quantinuum's Helios trapped-ion system, the same hardware we covered in our ",{"type":21,"tag":29,"props":1356,"children":1357},{"href":1141},[1358],{"type":26,"value":1359},"piece on Helios's logical qubit encoding",{"type":26,"value":534},{"type":21,"tag":59,"props":1362,"children":1364},{"id":1363},"read-the-claim-at-the-scale-it-was-tested",[1365],{"type":26,"value":1366},"Read the claim at the scale it was tested",{"type":21,"tag":22,"props":1368,"children":1369},{},[1370,1372,1378],{"type":26,"value":1371},"This result is about circuit generation, not about simulating a bigger molecule than anyone has before. It is a different kind of advance from the ",{"type":21,"tag":29,"props":1373,"children":1375},{"href":1374},"\u002Fblog\u002Fquantum-chemistry-protein-scale-2026",[1376],{"type":26,"value":1377},"Cleveland Clinic and IBM protein-scale simulation",{"type":26,"value":1379}," we covered separately, which pushed the size of the system being simulated. ADAPT-GQE instead pushes the speed and cost of building the circuit for a given molecule, tested here on one well-characterized drug compound. Whether the trained model generalizes to structurally different molecules without retraining, and how it performs as molecule size grows well past imipramine, are the open questions worth watching in follow-up work rather than assuming from this result alone.",{"type":21,"tag":59,"props":1381,"children":1383},{"id":1382},"why-this-fits-a-pattern-worth-naming",[1384],{"type":26,"value":1385},"Why this fits a pattern worth naming",{"type":21,"tag":22,"props":1387,"children":1388},{},[1389,1391,1396],{"type":26,"value":1390},"Pfizer's involvement gives this a real commercial angle instead of a research-only demonstration, and it points at a broader shift already visible across 2026 quantum chemistry news: less effort spent proving quantum computers touch a toy molecule, more effort spent removing the practical bottlenecks, molecule size in IBM's case, circuit design cost here, standing between a working method and something a pharmaceutical team would use. If you want to build the small-molecule version of this yourself on a free simulator, our ",{"type":21,"tag":29,"props":1392,"children":1393},{"href":893},[1394],{"type":26,"value":1395},"PennyLane VQE guide",{"type":26,"value":1397}," walks through the manual ansatz-construction approach ADAPT-GQE is trying to replace.",{"title":7,"searchDepth":335,"depth":335,"links":1399},[1400,1401,1402,1403],{"id":1330,"depth":335,"text":1333},{"id":1346,"depth":335,"text":1349},{"id":1363,"depth":335,"text":1366},{"id":1382,"depth":335,"text":1385},"content:blog:adapt-gqe-ai-quantum-circuit-design.md","blog\u002Fadapt-gqe-ai-quantum-circuit-design.md","blog\u002Fadapt-gqe-ai-quantum-circuit-design",{"_path":1408,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1409,"description":1410,"date":10,"author":11,"tags":1411,"readingTime":453,"body":1412,"_type":715,"_id":1478,"_source":717,"_file":1479,"_stem":1480,"_extension":720},"\u002Fblog\u002Fglobalfoundries-chips-silicon-photonics-award","GlobalFoundries Gets $300 Million to Build Silicon Photonics in the US","GlobalFoundries signed a letter of intent for a $300 million CHIPS Act award to scale US silicon photonics manufacturing. It is not a quantum computing grant, but the manufacturing capacity it builds matters for photonic quantum hardware and quantum networking too.",[1033,730],{"type":18,"children":1413,"toc":1473},[1414,1419,1425,1430,1436,1462,1468],{"type":21,"tag":22,"props":1415,"children":1416},{},[1417],{"type":26,"value":1418},"GlobalFoundries signed a letter of intent with the US Department of Commerce on July 29, 2026, for a $300 million CHIPS Act award to scale up US silicon photonics manufacturing. Read the headline carefully: this is not quantum computing funding. It is an AI-infrastructure story first. It belongs on this site because the manufacturing capacity it builds is the same capacity photonic quantum computing and quantum networking depend on.",{"type":21,"tag":59,"props":1420,"children":1422},{"id":1421},"what-the-award-funds",[1423],{"type":26,"value":1424},"What the award funds",{"type":21,"tag":22,"props":1426,"children":1427},{},[1428],{"type":26,"value":1429},"Silicon photonics moves data using light instead of electrical signals, which gives AI systems more bandwidth per watt of power drawn. The funding covers co-packaged optics, an approach that places photonic circuitry directly next to processors instead of routing signals through separate optical modules, along with next-generation optical materials, wafer technology, and advanced packaging. GlobalFoundries states target performance of 400 gigabits per second in throughput and energy consumption cut to a fifth of current-generation systems. The Department of Commerce receives roughly a 1 percent equity stake in GlobalFoundries as part of the deal. Work will run through GlobalFoundries' existing facilities in Malta, New York, and Burlington, Vermont.",{"type":21,"tag":59,"props":1431,"children":1433},{"id":1432},"why-a-chipmakers-ai-deal-shows-up-on-a-quantum-computing-site",[1434],{"type":26,"value":1435},"Why a chipmaker's AI deal shows up on a quantum computing site",{"type":21,"tag":22,"props":1437,"children":1438},{},[1439,1441,1446,1448,1453,1455,1461],{"type":26,"value":1440},"Much of today's silicon photonics and advanced optical packaging supply chain sits outside the United States, which is the same domestic-manufacturing motivation behind ",{"type":21,"tag":29,"props":1442,"children":1443},{"href":1117},[1444],{"type":26,"value":1445},"IonQ's acquisition of SkyWater Technology",{"type":26,"value":1447},". The connection to quantum computing is indirect but real: photonic quantum computing companies like Xanadu and PsiQuantum, tracked on our ",{"type":21,"tag":29,"props":1449,"children":1450},{"href":862},[1451],{"type":26,"value":1452},"industry landscape",{"type":26,"value":1454},", depend on exactly this kind of silicon photonics fabrication and packaging capacity. So do quantum networking efforts aimed at linking separate QPUs with optical interconnects, the subject of our piece on ",{"type":21,"tag":29,"props":1456,"children":1458},{"href":1457},"\u002Fblog\u002Fquantum-networking-distributed-computing",[1459],{"type":26,"value":1460},"quantum networking and distributed quantum computing",{"type":26,"value":534},{"type":21,"tag":59,"props":1463,"children":1465},{"id":1464},"what-this-does-and-does-not-tell-you-about-quantum-hardware",[1466],{"type":26,"value":1467},"What this does and does not tell you about quantum hardware",{"type":21,"tag":22,"props":1469,"children":1470},{},[1471],{"type":26,"value":1472},"This award funds general-purpose silicon photonics aimed at AI data centers, not a quantum-specific fabrication line. No quantum hardware vendor is named as a direct beneficiary in the announcement. What it does tell you is that the underlying manufacturing base for photonic and optically networked quantum hardware is getting real government investment and real domestic capacity, for reasons that have nothing to do with quantum computing at all. That is worth knowing if you are trying to judge how much of the \"domestic quantum supply chain\" story making the rounds in 2026 is quantum-specific policy versus a broader chip and AI-infrastructure push that quantum hardware happens to benefit from as a side effect.",{"title":7,"searchDepth":335,"depth":335,"links":1474},[1475,1476,1477],{"id":1421,"depth":335,"text":1424},{"id":1432,"depth":335,"text":1435},{"id":1464,"depth":335,"text":1467},"content:blog:globalfoundries-chips-silicon-photonics-award.md","blog\u002Fglobalfoundries-chips-silicon-photonics-award.md","blog\u002Fglobalfoundries-chips-silicon-photonics-award",{"_path":1482,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1483,"description":1484,"date":10,"author":11,"tags":1485,"readingTime":496,"body":1486,"_type":715,"_id":1613,"_source":717,"_file":1614,"_stem":1615,"_extension":720},"\u002Fblog\u002Fgoogle-willow-reinforcement-learning-calibration","Google Taught Willow to Calibrate Itself While It Runs Error Correction","Google Quantum AI and DeepMind published a Nature paper showing a reinforcement learning agent recalibrates the Willow processor using the same error-detection data error correction already produces, setting a new logical error rate record for the surface code.",[730,729,1033],{"type":18,"children":1487,"toc":1607},[1488,1510,1516,1521,1532,1538,1550,1555,1561,1566,1578,1584,1589],{"type":21,"tag":22,"props":1489,"children":1490},{},[1491,1493,1500,1501,1508],{"type":26,"value":1492},"Google Quantum AI and Google DeepMind published a paper in Nature on July 10, 2026 showing that a reinforcement learning agent keeps a quantum processor calibrated using data the processor was already producing, rather than pausing to run a separate tune-up routine, according to ",{"type":21,"tag":29,"props":1494,"children":1497},{"href":1495,"rel":1496},"https:\u002F\u002Fthequantuminsider.com\u002F2026\u002F07\u002F10\u002Fgoogle-study-shows-quantum-computer-can-learn-from-its-own-errors-while-it-computes\u002F",[44],[1498],{"type":26,"value":1499},"The Quantum Insider",{"type":26,"value":236},{"type":21,"tag":29,"props":1502,"children":1505},{"href":1503,"rel":1504},"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-026-10759-2",[44],[1506],{"type":26,"value":1507},"the paper itself",{"type":26,"value":1509},". The work ran on Google's Willow superconducting processor and was led by Volodymyr Sivak and Alexis Morvan, with 299 authors credited in total.",{"type":21,"tag":59,"props":1511,"children":1513},{"id":1512},"the-problem-this-solves",[1514],{"type":26,"value":1515},"The problem this solves",{"type":21,"tag":22,"props":1517,"children":1518},{},[1519],{"type":26,"value":1520},"Every quantum processor drifts. Control parameters, the microwave pulse amplitudes, frequencies, and coupling strengths that drive each qubit, slowly shift due to temperature changes, material aging, and electronic noise. The standard fix is periodic recalibration: pause the machine, run a dedicated tune-up sequence, then resume. That works, but it treats calibration and computation as separate activities competing for the same hardware time, and it only catches drift at the moments you choose to check for it.",{"type":21,"tag":22,"props":1522,"children":1523},{},[1524,1530],{"type":21,"tag":29,"props":1525,"children":1527},{"href":1526},"\u002Fglossary\u002Fquantum-error-correction",[1528],{"type":26,"value":1529},"Quantum error correction",{"type":26,"value":1531}," already generates a stream of error-detection events as a normal part of running any error-corrected circuit. Google's insight was that this data is not only useful for correcting the logical qubit's state. It also tells you, continuously, how well the underlying hardware is behaving, which is exactly the signal a calibration system needs.",{"type":21,"tag":59,"props":1533,"children":1535},{"id":1534},"what-the-reinforcement-learning-agent-does",[1536],{"type":26,"value":1537},"What the reinforcement learning agent does",{"type":21,"tag":22,"props":1539,"children":1540},{},[1541,1543,1548],{"type":26,"value":1542},"The system trains a reinforcement learning agent to read the error-detection events produced during normal error-correction cycles and use them to continuously adjust more than 1,000 control parameters, without stopping the computation to do it. Google tested the approach on distance-5 and distance-7 ",{"type":21,"tag":29,"props":1544,"children":1545},{"href":854},[1546],{"type":26,"value":1547},"surface codes",{"type":26,"value":1549}," and a distance-5 color code on the 105-qubit Willow chip, and ran simulations extending to a distance-15 surface code involving roughly 40,000 parameters, to check whether the approach holds up at a scale beyond what current hardware supports.",{"type":21,"tag":22,"props":1551,"children":1552},{},[1553],{"type":26,"value":1554},"The results at the tested scale: a 20% reduction in logical error rate beyond what conventional calibration achieves, even after exhaustive manual tuning. Under artificially injected hardware drift, meant to simulate the kind of degradation a real system experiences over hours of operation, the reinforcement learning approach cut the logical error rate by 24% and made performance 2.4 times more stable than static calibration. Adding decoder-parameter adaptation on top pushed that to a 31% error reduction and 3.5 times more stability. On the distance-7 surface code specifically, the system reached a logical error rate of 7.72 × 10⁻⁴ per cycle, which the authors report as a new record for that code family.",{"type":21,"tag":59,"props":1556,"children":1558},{"id":1557},"why-this-matters-more-than-the-specific-numbers",[1559],{"type":26,"value":1560},"Why this matters more than the specific numbers",{"type":21,"tag":22,"props":1562,"children":1563},{},[1564],{"type":26,"value":1565},"This is the first demonstration of reinforcement learning controlling error correction at the scale of a full error-corrected processor. Earlier experimental work applied similar reinforcement learning techniques to isolated gates or to bosonic codes, smaller, more contained systems. Running it across a full surface-code processor with over 1,000 live control parameters is a meaningfully harder engineering problem, and the fact that it improved on hand-tuned calibration rather than merely matching it is the part worth sitting with.",{"type":21,"tag":22,"props":1567,"children":1568},{},[1569,1571,1576],{"type":26,"value":1570},"Drift is a problem that gets worse, not better, as machines scale up. A processor with a few dozen qubits is recalibrated by hand often enough to stay ahead of drift. A processor with thousands of qubits, the scale ",{"type":21,"tag":29,"props":1572,"children":1573},{"href":1067},[1574],{"type":26,"value":1575},"IBM's roadmap",{"type":26,"value":1577}," and others are explicitly building toward, cannot rely on manual tune-up passes keeping pace with how many parameters need adjusting. A system that recalibrates itself continuously, using data the machine produces anyway, is a more plausible path to keeping a much larger processor stable than scaling up the current manual approach.",{"type":21,"tag":59,"props":1579,"children":1581},{"id":1580},"what-is-still-unproven",[1582],{"type":26,"value":1583},"What is still unproven",{"type":21,"tag":22,"props":1585,"children":1586},{},[1587],{"type":26,"value":1588},"This result was demonstrated on Google's own Willow hardware and published in a peer-reviewed venue, which puts it on firmer ground than a vendor press release. It does not yet tell you whether the same reinforcement learning approach transfers cleanly to a different qubit modality, a different code family beyond the ones tested, or a processor an order of magnitude larger than Willow's 105 qubits. The distance-15 simulation results are exactly that: simulations, not a demonstration on real hardware at that scale.",{"type":21,"tag":22,"props":1590,"children":1591},{},[1592,1594,1599,1600,1605],{"type":26,"value":1593},"What is established is narrower and still significant: on real hardware, at the scale tested, learning from the error-correction data a processor already generates beats static calibration, and does so by a wide enough margin to set a new record for the surface code. Our ",{"type":21,"tag":29,"props":1595,"children":1596},{"href":854},[1597],{"type":26,"value":1598},"error correction explainer",{"type":26,"value":236},{"type":21,"tag":29,"props":1601,"children":1602},{"href":751},[1603],{"type":26,"value":1604},"logical qubits and fault tolerance piece",{"type":26,"value":1606}," cover the broader context for why keeping a large processor calibrated is as hard a problem as building the qubits in the first place.",{"title":7,"searchDepth":335,"depth":335,"links":1608},[1609,1610,1611,1612],{"id":1512,"depth":335,"text":1515},{"id":1534,"depth":335,"text":1537},{"id":1557,"depth":335,"text":1560},{"id":1580,"depth":335,"text":1583},"content:blog:google-willow-reinforcement-learning-calibration.md","blog\u002Fgoogle-willow-reinforcement-learning-calibration.md","blog\u002Fgoogle-willow-reinforcement-learning-calibration",{"_path":1617,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1618,"description":1619,"date":10,"author":11,"tags":1620,"readingTime":474,"body":1621,"_type":715,"_id":1718,"_source":717,"_file":1719,"_stem":1720,"_extension":720},"\u002Fblog\u002Fhrl-self-operating-silicon-quantum-processor","HRL Built a Silicon Quantum Processor That Runs Its Own Error Correction","HRL Laboratories published a silicon quantum processor in Nature that runs error correction with a control chip inside the cryostat, cutting the room-temperature wiring that limits how many qubits a machine holds.",[730,729,1033],{"type":18,"children":1622,"toc":1713},[1623,1635,1641,1646,1664,1670,1675,1680,1686,1697],{"type":21,"tag":22,"props":1624,"children":1625},{},[1626,1628,1633],{"type":26,"value":1627},"Most quantum computers depend on racks of room-temperature electronics wired down into a cryostat, one connection per control signal, per qubit. That wiring, not qubit physics, is one of the real limits on how many qubits a machine holds. On July 29, 2026, HRL Laboratories published a result in Nature that attacks this problem directly: a silicon quantum processor with its own control chip living inside the cryostat, running ",{"type":21,"tag":29,"props":1629,"children":1630},{"href":1526},[1631],{"type":26,"value":1632},"error correction",{"type":26,"value":1634}," on its own.",{"type":21,"tag":59,"props":1636,"children":1638},{"id":1637},"what-hrl-built",[1639],{"type":26,"value":1640},"What HRL built",{"type":21,"tag":22,"props":1642,"children":1643},{},[1644],{"type":26,"value":1645},"The device holds 18 qubits. Next to them, inside the same cryostat, sits a custom CMOS controller operating at around minus 450 degrees Fahrenheit, close to the qubits themselves rather than at room temperature. That controller generates every signal the error-correction routine needs and runs the whole cycle with no real-time input from outside the cold chamber. A high-density ribbon cable, 296 channels of superconducting niobium on polyimide, connects the pieces while managing the heat load a control chip normally cannot tolerate at those temperatures.",{"type":21,"tag":22,"props":1647,"children":1648},{},[1649,1651,1656,1657,1662],{"type":26,"value":1650},"The controller chip itself was manufactured at a commercial foundry, the same kind of manufacturing step behind ",{"type":21,"tag":29,"props":1652,"children":1653},{"href":1117},[1654],{"type":26,"value":1655},"IonQ's recent purchase of SkyWater Technology",{"type":26,"value":236},{"type":21,"tag":29,"props":1658,"children":1659},{"href":1408},[1660],{"type":26,"value":1661},"GlobalFoundries' new silicon photonics funding",{"type":26,"value":1663},". None of these stories are about qubit physics. All three are about the unglamorous manufacturing and packaging work that decides whether a lab result becomes a machine you scale.",{"type":21,"tag":59,"props":1665,"children":1667},{"id":1666},"why-moving-the-controller-matters-more-than-it-sounds",[1668],{"type":26,"value":1669},"Why moving the controller matters more than it sounds",{"type":21,"tag":22,"props":1671,"children":1672},{},[1673],{"type":26,"value":1674},"Every wire running from room temperature into a cryostat carries heat with it, and heat is the enemy of a qubit that needs to stay near absolute zero. Today's largest machines manage this by keeping wiring counts down, which caps how many qubits fit on one system before the cooling and wiring overhead becomes the bottleneck rather than the qubits themselves. A control chip that lives in the cold zone and runs error correction locally removes a big share of that wiring, one wired connection per function instead of one per qubit.",{"type":21,"tag":22,"props":1676,"children":1677},{},[1678],{"type":26,"value":1679},"That is a real, checkable engineering claim, and it is a different kind of claim than a qubit count or a fidelity number. It says nothing about how good HRL's 18 qubits are compared to a superconducting or trapped-ion qubit elsewhere. It says something about whether a silicon-based approach scales its control infrastructure without the wiring problem growing linearly with qubit count.",{"type":21,"tag":59,"props":1681,"children":1683},{"id":1682},"where-this-sits-next-to-other-2026-scaling-work",[1684],{"type":26,"value":1685},"Where this sits next to other 2026 scaling work",{"type":21,"tag":22,"props":1687,"children":1688},{},[1689,1695],{"type":21,"tag":29,"props":1690,"children":1692},{"href":1691},"\u002Fblog\u002Fwarwick-quantum-phononic-links",[1693],{"type":26,"value":1694},"Warwick researchers published a different answer",{"type":26,"value":1696}," to a related scaling problem the same week: how to let distant qubits on one chip communicate at all, rather than only their nearest neighbors. Read together, the two results point at the same underlying truth. Qubit count and gate fidelity get the headlines, but wiring, control electronics, and qubit-to-qubit communication are the problems that decide whether a design reaches a million qubits or stalls in the hundreds.",{"type":21,"tag":22,"props":1698,"children":1699},{},[1700,1702,1707,1708,1712],{"type":26,"value":1701},"HRL's result is a working 18-qubit device with a published, peer-reviewed result behind it, not a concept paper. That distinction is worth holding onto as you read coverage of scaling claims generally. A demonstrated device at a small scale is a different kind of evidence than a proposed mechanism aimed at a much larger one, even when both are genuine contributions. For more on how silicon-based qubits compare to the trapped-ion, superconducting, and photonic approaches competing for the same scaling problem, see our ",{"type":21,"tag":29,"props":1703,"children":1704},{"href":806},[1705],{"type":26,"value":1706},"hardware overview",{"type":26,"value":236},{"type":21,"tag":29,"props":1709,"children":1710},{"href":862},[1711],{"type":26,"value":1452},{"type":26,"value":534},{"title":7,"searchDepth":335,"depth":335,"links":1714},[1715,1716,1717],{"id":1637,"depth":335,"text":1640},{"id":1666,"depth":335,"text":1669},{"id":1682,"depth":335,"text":1685},"content:blog:hrl-self-operating-silicon-quantum-processor.md","blog\u002Fhrl-self-operating-silicon-quantum-processor.md","blog\u002Fhrl-self-operating-silicon-quantum-processor",{"_path":1722,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1723,"description":1724,"date":10,"author":11,"tags":1725,"readingTime":496,"body":1726,"_type":715,"_id":1955,"_source":717,"_file":1956,"_stem":1957,"_extension":720},"\u002Fblog\u002Fibm-hrl-laboratories-acquisition-explained","IBM Is Acquiring HRL Laboratories: What the Silicon Spin Qubit Deal Adds","IBM has signed a definitive agreement to acquire HRL Laboratories from Boeing and General Motors, days before HRL's own paper on a self-correcting 18-qubit silicon processor. Here is what the deal covers and what it does not.",[1033,730,729],{"type":18,"children":1727,"toc":1948},[1728,1742,1754,1760,1787,1807,1812,1818,1823,1828,1833,1839,1844,1849,1855,1860,1865,1871,1876,1892],{"type":21,"tag":22,"props":1729,"children":1730},{},[1731,1733,1740],{"type":26,"value":1732},"IBM announced on July 23, 2026 that it signed a definitive agreement to acquire HRL Laboratories, a research institution jointly owned by Boeing and General Motors, according to ",{"type":21,"tag":29,"props":1734,"children":1737},{"href":1735,"rel":1736},"https:\u002F\u002Fnewsroom.ibm.com\u002F2026-07-23-ibm-to-acquire-hrl-laboratories-to-power-the-future-of-quantum",[44],[1738],{"type":26,"value":1739},"IBM's newsroom announcement",{"type":26,"value":1741},". Deal terms were not disclosed. The transaction is expected to close by the end of the third quarter of 2026, subject to regulatory approval. Boeing and GM will continue to partner with IBM on quantum applications after the deal closes.",{"type":21,"tag":22,"props":1743,"children":1744},{},[1745,1747,1752],{"type":26,"value":1746},"Six days later, HRL published the result that explains why IBM wanted the lab: a silicon quantum processor that runs its own error correction, ",{"type":21,"tag":29,"props":1748,"children":1749},{"href":1617},[1750],{"type":26,"value":1751},"which we covered here",{"type":26,"value":1753},". The timing is not a coincidence worth over-reading, since large acquisitions and peer-reviewed papers run on separate clocks, but it does mean the acquisition announcement and the technical proof point landed in the same week.",{"type":21,"tag":59,"props":1755,"children":1757},{"id":1756},"what-hrl-brings",[1758],{"type":26,"value":1759},"What HRL brings",{"type":21,"tag":22,"props":1761,"children":1762},{},[1763,1765,1772,1774,1779,1781,1786],{"type":26,"value":1764},"HRL specializes in silicon-spin qubit engineering, quantum sensing, and quantum materials, according to ",{"type":21,"tag":29,"props":1766,"children":1769},{"href":1767,"rel":1768},"https:\u002F\u002Fresearch.ibm.com\u002Fblog\u002Fhrl-laboratories-ibm",[44],[1770],{"type":26,"value":1771},"IBM Research's own writeup",{"type":26,"value":1773},". That is a different qubit modality from IBM's existing superconducting transmon platform, the one behind ",{"type":21,"tag":29,"props":1775,"children":1776},{"href":1067},[1777],{"type":26,"value":1778},"Nighthawk",{"type":26,"value":1780}," and the ",{"type":21,"tag":29,"props":1782,"children":1783},{"href":1274},[1784],{"type":26,"value":1785},"Starling and Blue Jay roadmap",{"type":26,"value":534},{"type":21,"tag":22,"props":1788,"children":1789},{},[1790,1792,1798,1800,1805],{"type":26,"value":1791},"Spin qubits encode information in the spin state of individual electrons held in quantum dots, tiny confined regions on a silicon chip. IBM Research describes HRL's most recent processor as 56 quantum dots configured to run as 18 ",{"type":21,"tag":29,"props":1793,"children":1795},{"href":1794},"\u002Fglossary\u002Fqubit",[1796],{"type":26,"value":1797},"qubits",{"type":26,"value":1799},", with a CMOS control chip inside the cryostat handling error correction locally instead of routing every signal through room-temperature electronics. Our ",{"type":21,"tag":29,"props":1801,"children":1802},{"href":1617},[1803],{"type":26,"value":1804},"earlier piece on that result",{"type":26,"value":1806}," goes into why moving the controller into the cold zone matters: it cuts the wiring count that otherwise limits how many qubits a machine holds.",{"type":21,"tag":22,"props":1808,"children":1809},{},[1810],{"type":26,"value":1811},"The pitch for spin qubits next to superconducting ones is manufacturing, not raw performance. Both approaches need cryogenic cooling, and both are built with silicon fabrication techniques closer to conventional chipmaking than to anything exotic. Spin qubits also occupy a smaller physical footprint per qubit than superconducting circuits, at least in principle, since a quantum dot is a nanometer-scale feature rather than a micron-scale resonator. Whether that footprint advantage survives contact with a full error-corrected system at scale is exactly the kind of claim this site treats skeptically until independent results confirm it. A demonstrated 18-qubit device is real progress. It is not yet evidence about what a 10,000-qubit spin system costs to build or run.",{"type":21,"tag":59,"props":1813,"children":1815},{"id":1814},"why-ibm-is-buying-instead-of-partnering",[1816],{"type":26,"value":1817},"Why IBM is buying instead of partnering",{"type":21,"tag":22,"props":1819,"children":1820},{},[1821],{"type":26,"value":1822},"IBM already runs the industry's broadest quantum platform on superconducting hardware. Betting on a second qubit modality by acquisition rather than an ordinary research partnership signals that IBM wants the engineering team and the intellectual property inside its own roadmap, not only access to a paper.",{"type":21,"tag":22,"props":1824,"children":1825},{},[1826],{"type":26,"value":1827},"This deal is not IBM's first move in that direction this year. In May 2026, IBM and the U.S. Department of Commerce announced Anderon, described as the first purpose-built quantum wafer foundry in the country, a 300-millimeter facility in Albany, New York, backed by a proposed $1 billion CHIPS Act award alongside $1 billion in IBM's own cash. Then on June 2, 2026, IBM committed more than $10 billion to quantum computing over five years, spanning research, manufacturing, ecosystem partnerships, and explicitly, mergers and acquisitions. The HRL deal, announced seven weeks later, is that acquisitions line item turning into an actual transaction.",{"type":21,"tag":22,"props":1829,"children":1830},{},[1831],{"type":26,"value":1832},"Read together, the sequence looks like a company building the manufacturing base first (Anderon), backing it with capital (the $10 billion commitment), and then acquiring a qubit modality it did not already have in-house (HRL). None of that guarantees the bet pays off. It does mean the acquisition fits a pattern rather than arriving as an isolated headline.",{"type":21,"tag":59,"props":1834,"children":1836},{"id":1835},"the-people-making-the-case",[1837],{"type":26,"value":1838},"The people making the case",{"type":21,"tag":22,"props":1840,"children":1841},{},[1842],{"type":26,"value":1843},"Jay Gambetta, IBM's Director of Research and an IBM Fellow, framed the deal as HRL helping IBM \"advance toward the frontiers of quantum innovation\" and strengthening its \"long-term plans to deliver quantum computing, sensing and networking advances,\" per IBM's announcement. HRL's president and CEO, Rob Vasquez, called the move the \"natural next chapter\" for a team that has spent years exploring how future quantum computers might be built at scales that seem impossible today.",{"type":21,"tag":22,"props":1845,"children":1846},{},[1847],{"type":26,"value":1848},"Those are the kinds of quotes every acquisition announcement produces, and they are worth reading as exactly that: the framing both companies agreed to put in front of the press, not an independent assessment. The parts of this deal that are checkable are the corporate facts (signed agreement, expected close date, the ownership change from Boeing and GM to IBM) and HRL's public research record. The parts that are not yet checkable are whether silicon-spin qubits reach useful scale faster inside IBM than they would have independently.",{"type":21,"tag":59,"props":1850,"children":1852},{"id":1851},"hrls-history-is-longer-than-the-quantum-story-suggests",[1853],{"type":26,"value":1854},"HRL's history is longer than the quantum story suggests",{"type":21,"tag":22,"props":1856,"children":1857},{},[1858],{"type":26,"value":1859},"HRL traces back to Hughes Research Laboratories, founded by Howard Hughes in 1948. Its best-known achievement predates quantum computing by six decades: Theodore Maiman built the world's first working laser at the lab's Malibu facility on May 16, 1960, using a synthetic ruby crystal. The lab reorganized as HRL Laboratories, LLC in 1997, after Hughes Aircraft's ownership was restructured, and has operated since under joint ownership by Boeing and General Motors, serving both commercial and U.S. government customers across sensing, communications, advanced manufacturing, and materials science, not only quantum computing.",{"type":21,"tag":22,"props":1861,"children":1862},{},[1863],{"type":26,"value":1864},"That history matters for one reason: HRL is an established, credentialed lab with a long track record of peer-reviewed, independently reproduced results, not a startup selling a roadmap. When it publishes a Nature paper, as it did on the self-operating silicon processor, the result carries the weight of that track record. That is a different kind of credibility than a press release, and it is worth distinguishing the two when reading coverage of this acquisition.",{"type":21,"tag":59,"props":1866,"children":1868},{"id":1867},"what-to-watch-once-the-deal-closes",[1869],{"type":26,"value":1870},"What to watch once the deal closes",{"type":21,"tag":22,"props":1872,"children":1873},{},[1874],{"type":26,"value":1875},"The close itself, expected by the end of Q3 2026, is the first checkable milestone. After that, watch whether IBM folds HRL's spin-qubit work into a named point on its public roadmap, the way Starling and Blue Jay are already named and dated, or keeps it as a longer-horizon research bet without a committed delivery date. Also watch whether Anderon's foundry plans expand beyond superconducting wafers to cover spin-qubit fabrication, since that would be the concrete sign that IBM intends to manufacture both modalities at scale rather than run HRL as a separate research track.",{"type":21,"tag":22,"props":1877,"children":1878},{},[1879,1881,1885,1886,1890],{"type":26,"value":1880},"For now, the honest summary is narrower than the headlines: IBM bought a credentialed silicon-spin qubit research team and its intellectual property, for an undisclosed price, weeks after committing capital to exactly this kind of move. The technology bet is real. Whether it changes IBM's timeline to a useful quantum computer is not something this announcement, or HRL's paper, answers on its own. Our ",{"type":21,"tag":29,"props":1882,"children":1883},{"href":806},[1884],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":1887,"children":1888},{"href":862},[1889],{"type":26,"value":1452},{"type":26,"value":1891}," are the places to track how this compares against the trapped-ion, photonic, and neutral-atom approaches other companies are betting on instead.",{"type":21,"tag":22,"props":1893,"children":1894},{},[1895,1900,1902,1908,1909,1915,1916,1923,1924,1931,1932,1939,1940,1947],{"type":21,"tag":71,"props":1896,"children":1897},{},[1898],{"type":26,"value":1899},"Sources:",{"type":26,"value":1901}," ",{"type":21,"tag":29,"props":1903,"children":1905},{"href":1735,"rel":1904},[44],[1906],{"type":26,"value":1907},"IBM newsroom announcement",{"type":26,"value":367},{"type":21,"tag":29,"props":1910,"children":1912},{"href":1767,"rel":1911},[44],[1913],{"type":26,"value":1914},"IBM Research blog",{"type":26,"value":367},{"type":21,"tag":29,"props":1917,"children":1920},{"href":1918,"rel":1919},"https:\u002F\u002Fsiliconangle.com\u002F2026\u002F07\u002F23\u002Fibm-acquires-quantum-computing-research-lab-hrl\u002F",[44],[1921],{"type":26,"value":1922},"SiliconANGLE",{"type":26,"value":367},{"type":21,"tag":29,"props":1925,"children":1928},{"href":1926,"rel":1927},"https:\u002F\u002Fwww.govconwire.com\u002Farticles\u002Fibm-buy-hrl-laboratories-quantum-computing",[44],[1929],{"type":26,"value":1930},"GovConWire",{"type":26,"value":367},{"type":21,"tag":29,"props":1933,"children":1936},{"href":1934,"rel":1935},"https:\u002F\u002Fnewsroom.ibm.com\u002F2026-06-02-ibm-commits-more-than-10-billion-to-quantum-computing,-funding-its-roadmap-from-todays-leading-systems-to-the-worlds-first-fault-tolerant-quantum-computers",[44],[1937],{"type":26,"value":1938},"IBM's $10 billion quantum investment announcement",{"type":26,"value":367},{"type":21,"tag":29,"props":1941,"children":1944},{"href":1942,"rel":1943},"https:\u002F\u002Fwww.hrl.com\u002Fabout\u002Fhistory",[44],[1945],{"type":26,"value":1946},"HRL Laboratories history",{"type":26,"value":534},{"title":7,"searchDepth":335,"depth":335,"links":1949},[1950,1951,1952,1953,1954],{"id":1756,"depth":335,"text":1759},{"id":1814,"depth":335,"text":1817},{"id":1835,"depth":335,"text":1838},{"id":1851,"depth":335,"text":1854},{"id":1867,"depth":335,"text":1870},"content:blog:ibm-hrl-laboratories-acquisition-explained.md","blog\u002Fibm-hrl-laboratories-acquisition-explained.md","blog\u002Fibm-hrl-laboratories-acquisition-explained",{"_path":1067,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":1959,"description":1960,"date":10,"author":11,"tags":1961,"readingTime":16,"body":1963,"_type":715,"_id":2112,"_source":717,"_file":2113,"_stem":2114,"_extension":720},"IBM Claimed Quantum Advantage Three Times in One Day: What Each Result Shows","On July 30, 2026, IBM announced three separate quantum advantage results with three different partners. Here is what each one demonstrated, and why the verification method matters more than the headline.",[1033,730,1962],"Qiskit",{"type":18,"children":1964,"toc":2105},[1965,1970,1976,1989,2001,2007,2012,2017,2023,2028,2033,2039,2060,2065,2071,2076],{"type":21,"tag":22,"props":1966,"children":1967},{},[1968],{"type":26,"value":1969},"On July 30, 2026, IBM announced three quantum advantage results in a single day. Three different partners. Three different physics problems. Three different verification methods. Notice that pattern before you read any single headline. IBM Research director Jay Gambetta called this \"the quantum advantage era.\" Treat that claim the way we treat every vendor headline on this site. Take the pieces apart. Check each one. Decide what it means for you.",{"type":21,"tag":59,"props":1971,"children":1973},{"id":1972},"result-one-ibm-and-the-university-of-chicago-logical-qubits-at-scale",[1974],{"type":26,"value":1975},"Result one: IBM and the University of Chicago, logical qubits at scale",{"type":21,"tag":22,"props":1977,"children":1978},{},[1979,1981,1987],{"type":26,"value":1980},"The team ran an encoded circuit using 70 ",{"type":21,"tag":29,"props":1982,"children":1984},{"href":1983},"\u002Fglossary\u002Flogical-qubit",[1985],{"type":26,"value":1986},"logical qubits",{"type":26,"value":1988}," through thousands of logical operations, with a logical error rate roughly 10 times lower than the physical error rate underneath it. The task finished in about 15 minutes on IBM hardware. Leading classical simulation methods faced runtimes described as prohibitive for the same task.",{"type":21,"tag":22,"props":1990,"children":1991},{},[1992,1994,1999],{"type":26,"value":1993},"Notice what the coverage did not state: the physical qubit count behind those 70 logical qubits. Our ",{"type":21,"tag":29,"props":1995,"children":1996},{"href":751},[1997],{"type":26,"value":1998},"companion piece on logical qubits and fault tolerance",{"type":26,"value":2000}," covers why that ratio, not the logical qubit count alone, decides whether a result is efficient or brute-forced. Until that number surfaces, read \"70 logical qubits\" as a real result with one open question attached.",{"type":21,"tag":59,"props":2002,"children":2004},{"id":2003},"result-two-ibm-and-qedma-a-real-physics-question-checked-two-ways",[2005],{"type":26,"value":2006},"Result two: IBM and Qedma, a real physics question checked two ways",{"type":21,"tag":22,"props":2008,"children":2009},{},[2010],{"type":26,"value":2011},"This team simulated a 74-qubit two-dimensional Floquet Ising model on IBM Heron processors, using Qedma's QESEM error-mitigation software. RIKEN contributed, and BlueQubit supported the study. The team compared results against classical simulation methods, including Japan's Fugaku supercomputer, and found the quantum results held steady past the point where classical approaches stopped agreeing with each other.",{"type":21,"tag":22,"props":2013,"children":2014},{},[2015],{"type":26,"value":2016},"The trust-building step here is the part worth remembering. Two independent error-mitigation estimators, one heuristic and one rigorous, agreed with each other. The team then partially reproduced the experiment on a Quantinuum system, a competitor's hardware, and got consistent results. Cross-platform agreement is a stronger form of evidence than one vendor reporting a number from its own machine.",{"type":21,"tag":59,"props":2018,"children":2020},{"id":2019},"result-three-ibm-and-algorithmiq-checking-an-answer-nothing-else-verifies",[2021],{"type":26,"value":2022},"Result three: IBM and Algorithmiq, checking an answer nothing else verifies",{"type":21,"tag":22,"props":2024,"children":2025},{},[2026],{"type":26,"value":2027},"This is the hardest version of the problem. When no classical computer verifies a result at all, how do you know the quantum answer is right? IBM and Algorithmiq simulated a heterogeneous quantum material on IBM Heron hardware and built their case on noise modeling, error mitigation, and cross-platform testing rather than a classical ground truth.",{"type":21,"tag":22,"props":2029,"children":2030},{},[2031],{"type":26,"value":2032},"Algorithmiq also open-sourced a classical simulation package called monoprop, built specifically so outside researchers independently test future advantage claims instead of taking a press release at its word. That release matters more than it might look. It is the one part of this announcement any reader of this site is able to check directly.",{"type":21,"tag":59,"props":2034,"children":2036},{"id":2035},"why-the-trust-framing-matters-more-than-the-speed-framing",[2037],{"type":26,"value":2038},"Why the trust framing matters more than the speed framing",{"type":21,"tag":22,"props":2040,"children":2041},{},[2042,2044,2050,2052,2058],{"type":26,"value":2043},"\"Beyond classical simulation\" is not a new claim. Google made a related claim in 2019 with its ",{"type":21,"tag":29,"props":2045,"children":2047},{"href":2046},"\u002Fresearch\u002Fgoogle-quantum-supremacy-2019",[2048],{"type":26,"value":2049},"quantum supremacy result",{"type":26,"value":2051},", and better classical algorithms later closed part of that gap. IBM made a related but weaker claim in 2023 with ",{"type":21,"tag":29,"props":2053,"children":2055},{"href":2054},"\u002Fresearch\u002Fibm-quantum-utility-2023",[2056],{"type":26,"value":2057},"quantum utility",{"type":26,"value":2059},", which did not require beating classical computation outright, only being useful and comparable to it.",{"type":21,"tag":22,"props":2061,"children":2062},{},[2063],{"type":26,"value":2064},"This round targets a harder bar than either: not only \"we finished faster than classical methods,\" but \"here is independently checkable evidence for why you should trust the answer.\" That second half, the verification method, is the actual news in all three results. The raw qubit and gate counts are the part every previous announcement already led with.",{"type":21,"tag":59,"props":2066,"children":2068},{"id":2067},"what-happens-next",[2069],{"type":26,"value":2070},"What happens next",{"type":21,"tag":22,"props":2072,"children":2073},{},[2074],{"type":26,"value":2075},"Peer review and independent replication are the steps that eventually narrowed Google's 2019 claim, and they are the steps to watch for here too. Three same-day press announcements are not the same thing as three peer-reviewed, independently replicated results. Given Algorithmiq's open monoprop release, at least one path to that independent check already exists. Watch whether other research groups pick it up and whether their numbers match.",{"type":21,"tag":22,"props":2077,"children":2078},{},[2079,2081,2085,2086,2090,2092,2097,2098,2103],{"type":26,"value":2080},"For a broader sense of how IBM's hardware and error-mitigation approach compares with competing modalities, see our ",{"type":21,"tag":29,"props":2082,"children":2083},{"href":806},[2084],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":2087,"children":2088},{"href":862},[2089],{"type":26,"value":1452},{"type":26,"value":2091},". For the mechanics behind logical qubits and why overhead numbers matter more than headline qubit counts, our ",{"type":21,"tag":29,"props":2093,"children":2094},{"href":751},[2095],{"type":26,"value":2096},"fault tolerance explainer",{"type":26,"value":236},{"type":21,"tag":29,"props":2099,"children":2100},{"href":1141},[2101],{"type":26,"value":2102},"Quantinuum Helios piece",{"type":26,"value":2104}," cover the same questions from a different vendor's claims.",{"title":7,"searchDepth":335,"depth":335,"links":2106},[2107,2108,2109,2110,2111],{"id":1972,"depth":335,"text":1975},{"id":2003,"depth":335,"text":2006},{"id":2019,"depth":335,"text":2022},{"id":2035,"depth":335,"text":2038},{"id":2067,"depth":335,"text":2070},"content:blog:ibm-quantum-advantage-triple-announcement-2026.md","blog\u002Fibm-quantum-advantage-triple-announcement-2026.md","blog\u002Fibm-quantum-advantage-triple-announcement-2026",{"_path":2116,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":2117,"description":2118,"date":10,"author":11,"tags":2119,"readingTime":474,"body":2120,"_type":715,"_id":2235,"_source":717,"_file":2236,"_stem":2237,"_extension":720},"\u002Fblog\u002Finfleqtion-sqale-illinois-quantum-park","Infleqtion Will Put a 50-Logical-Qubit Machine in Chicago by 2027","Infleqtion is deploying Sqale, a neutral-atom quantum computer targeting 50 logical qubits, at the Illinois Quantum and Microelectronics Park in 2027. The company is also opening a Chicago center aimed specifically at energy-grid optimization problems.",[730,1033],{"type":18,"children":2121,"toc":2230},[2122,2143,2149,2167,2187,2193,2198,2203,2209,2214],{"type":21,"tag":22,"props":2123,"children":2124},{},[2125,2127,2134,2135,2141],{"type":26,"value":2126},"On July 21, 2026, Infleqtion announced it will deploy a fault-tolerant, neutral-atom quantum computer named Sqale at the Illinois Quantum and Microelectronics Park (IQMP), with hardware delivery planned for 2027, according to ",{"type":21,"tag":29,"props":2128,"children":2131},{"href":2129,"rel":2130},"https:\u002F\u002Finfleqtion.com\u002Finfleqtion-to-deploy-fault-tolerant-neutral-atom-quantum-computer-in-illinois\u002F",[44],[2132],{"type":26,"value":2133},"the company's own announcement",{"type":26,"value":236},{"type":21,"tag":29,"props":2136,"children":2139},{"href":2137,"rel":2138},"https:\u002F\u002Fthequantuminsider.com\u002F2026\u002F07\u002F23\u002Finfleqtion-neutral-atom-quantum-computer-illinois\u002F",[44],[2140],{"type":26,"value":1499},{"type":26,"value":2142},". What makes this worth a post rather than a routine roadmap update is how specific the commitment is: a named site, a named delivery year, and a named application area the company is building toward, rather than a general statement about future ambitions.",{"type":21,"tag":59,"props":2144,"children":2146},{"id":2145},"what-sqale-is-supposed-to-do",[2147],{"type":26,"value":2148},"What Sqale is supposed to do",{"type":21,"tag":22,"props":2150,"children":2151},{},[2152,2154,2158,2160,2165],{"type":26,"value":2153},"Sqale is designed to demonstrate more than 50 ",{"type":21,"tag":29,"props":2155,"children":2156},{"href":1983},[2157],{"type":26,"value":1986},{"type":26,"value":2159}," at launch, with a stated path to 100, built on a physical architecture Infleqtion says scales beyond 1,000 physical qubits. Like other ",{"type":21,"tag":29,"props":2161,"children":2162},{"href":806},[2163],{"type":26,"value":2164},"neutral-atom systems",{"type":26,"value":2166},", it uses lasers to trap and control individual atoms, and Infleqtion says the system will couple with NVIDIA's NVQLink for low-latency communication with classical GPU hardware, a pairing several vendors are now building toward as quantum systems increasingly need fast classical co-processing for error correction and control.",{"type":21,"tag":22,"props":2168,"children":2169},{},[2170,2172,2177,2179,2185],{"type":26,"value":2171},"Fifty logical qubits would be a meaningful jump if it holds up. For comparison, ",{"type":21,"tag":29,"props":2173,"children":2174},{"href":1141},[2175],{"type":26,"value":2176},"Quantinuum's Helios system claims 48 logical qubits",{"type":26,"value":2178}," from roughly 96 physical qubits, and ",{"type":21,"tag":29,"props":2180,"children":2182},{"href":2181},"\u002Fblog\u002Ftop-quantum-computing-companies-2026",[2183],{"type":26,"value":2184},"QuEra reported 96 logical qubits",{"type":26,"value":2186}," on 448 physical qubits in a January 2026 Nature paper. Sqale's ratio, more than 50 logical qubits on an architecture scaling past 1,000 physical qubits, would sit at a notably higher overhead than either of those results, at least based on the numbers disclosed so far. None of these figures are directly comparable without knowing the exact error rates and code distances each system targets, and Sqale's numbers are a 2027 target, not a measured result.",{"type":21,"tag":59,"props":2188,"children":2190},{"id":2189},"the-energy-grid-angle-is-the-more-interesting-part",[2191],{"type":26,"value":2192},"The energy grid angle is the more interesting part",{"type":21,"tag":22,"props":2194,"children":2195},{},[2196],{"type":26,"value":2197},"Infleqtion is also opening a Chicago Quantum Innovation Center focused specifically on applying quantum optimization to the electrical grid: unit commitment (deciding which power plants run when), contingency analysis (planning for equipment failures), and nuclear fuel loading. Partners named in the announcement include the National Quantum Algorithm Center, University of Chicago professor Fred Chong, Constellation Energy, and EPRI, alongside existing ARPA-E funding for a related project called ENCODE.",{"type":21,"tag":22,"props":2199,"children":2200},{},[2201],{"type":26,"value":2202},"CTO Pranav Gokhale framed the choice deliberately: \"Energy is one of the best proving grounds for quantum computing because the problems are consequential and immediate.\" That is a more specific and checkable claim than most quantum-for-industry pitches, since grid optimization problems already have well-defined classical benchmarks. If a quantum approach beats those benchmarks on a real utility's data, that is a concrete result. If it does not, that will also be visible, which is more accountability than most quantum use-case announcements carry.",{"type":21,"tag":59,"props":2204,"children":2206},{"id":2205},"what-is-confirmed-and-what-is-not",[2207],{"type":26,"value":2208},"What is confirmed and what is not",{"type":21,"tag":22,"props":2210,"children":2211},{},[2212],{"type":26,"value":2213},"The site, the delivery year, and the named partners are concrete facts. The 50-logical-qubit target and the path to 100 are Infleqtion's own projections for a system that has not shipped yet. Infleqtion does have a track record to weigh those projections against: the company already operates the only 100-physical-qubit neutral-atom system at the UK's National Quantum Computing Centre, plus a deployment in Japan, so this is a company with working hardware in the field already, not a pre-revenue startup making a first claim.",{"type":21,"tag":22,"props":2215,"children":2216},{},[2217,2219,2223,2224,2228],{"type":26,"value":2218},"CEO Matthew Kinsella called the deployment \"a landmark moment for quantum computing in Chicago,\" which is the kind of framing every regional deployment announcement uses about itself. The part worth tracking independently of that framing is whether Sqale ships on the stated 2027 timeline, and whether the energy-grid partnership produces a published result against a real utility's optimization problem rather than a simulated one. Our ",{"type":21,"tag":29,"props":2220,"children":2221},{"href":806},[2222],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":2225,"children":2226},{"href":862},[2227],{"type":26,"value":1452},{"type":26,"value":2229}," track how neutral-atom systems like this one compare against the trapped-ion, superconducting, and photonic platforms other companies are betting on.",{"title":7,"searchDepth":335,"depth":335,"links":2231},[2232,2233,2234],{"id":2145,"depth":335,"text":2148},{"id":2189,"depth":335,"text":2192},{"id":2205,"depth":335,"text":2208},"content:blog:infleqtion-sqale-illinois-quantum-park.md","blog\u002Finfleqtion-sqale-illinois-quantum-park.md","blog\u002Finfleqtion-sqale-illinois-quantum-park",{"_path":1117,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":2239,"description":2240,"date":10,"author":11,"tags":2241,"readingTime":496,"body":2242,"_type":715,"_id":2388,"_source":717,"_file":2389,"_stem":2390,"_extension":720},"IonQ Bought a $1.8 Billion Semiconductor Foundry: Here Is What Changes","IonQ completed its acquisition of SkyWater Technology, the largest US-based semiconductor foundry, and calls itself the only vertically integrated full-stack quantum platform company. Here is what that claim means and what it does not.",[1033,730,731],{"type":18,"children":2243,"toc":2380},[2244,2258,2264,2269,2275,2280,2285,2290,2296,2301,2319,2325,2330,2336,2355,2361],{"type":21,"tag":22,"props":2245,"children":2246},{},[2247,2249,2256],{"type":26,"value":2248},"IonQ completed its acquisition of SkyWater Technology on July 31, 2026. SkyWater is the largest semiconductor foundry operating entirely inside the United States. IonQ announced the deal in January 2026. Total equity value: roughly ",{"type":21,"tag":29,"props":2250,"children":2253},{"href":2251,"rel":2252},"https:\u002F\u002Finvestors.ionq.com\u002Fnews\u002Fnews-details\u002F2026\u002FIonQ-Completes-Acquisition-of-SkyWater-Technology\u002Fdefault.aspx",[44],[2254],{"type":26,"value":2255},"$1.8 billion",{"type":26,"value":2257},". SkyWater shareholders got $15.00 in cash plus IonQ stock, subject to a collar, for each share held. IonQ calls the combined company the only vertically integrated full-stack quantum platform company. That claim is specific and checkable. Read the parts separately before you accept the headline.",{"type":21,"tag":59,"props":2259,"children":2261},{"id":2260},"what-skywater-builds",[2262],{"type":26,"value":2263},"What SkyWater builds",{"type":21,"tag":22,"props":2265,"children":2266},{},[2267],{"type":26,"value":2268},"SkyWater does not chase the newest logic node against TSMC or Samsung. Its specialty is foundational semiconductor nodes and advanced packaging: mature process technology, not leading edge. Facilities sit in Minnesota, Florida, and Texas. Customers include commercial buyers and federal defense agencies. This distinction decides whether the acquisition targets what IonQ's hardware needs.",{"type":21,"tag":59,"props":2270,"children":2272},{"id":2271},"the-part-of-ionqs-stack-this-touches",[2273],{"type":26,"value":2274},"The part of IonQ's stack this touches",{"type":21,"tag":22,"props":2276,"children":2277},{},[2278],{"type":26,"value":2279},"Most coverage of this deal skipped one detail. A trapped-ion qubit is a single atom held in an electromagnetic field. Nobody fabricates the qubit. Ytterbium and barium ions occur naturally. They are identical to each other. No factory improves them. So what does a semiconductor foundry have to do with trapped-ion hardware?",{"type":21,"tag":22,"props":2281,"children":2282},{},[2283],{"type":26,"value":2284},"Everything around the ion. The ion trap chip is a microfabricated electrode structure that confines and shuttles ions across a surface. It is a real semiconductor device. So are the photonic components used for laser addressing and readout, and the cryogenic and RF control electronics that drive the whole system. None of that needs a 3-nanometer process. It needs the mature-node fabrication and advanced packaging SkyWater already sells to other industries.",{"type":21,"tag":22,"props":2286,"children":2287},{},[2288],{"type":26,"value":2289},"Read that way, the acquisition targets a real supply chain need. IonQ is pulling a piece of its existing supply chain in-house instead of buying it from a third party.",{"type":21,"tag":59,"props":2291,"children":2293},{"id":2292},"what-vertically-integrated-means-precisely",[2294],{"type":26,"value":2295},"What \"vertically integrated\" means, precisely",{"type":21,"tag":22,"props":2297,"children":2298},{},[2299],{"type":26,"value":2300},"No other quantum hardware company owns its foundry outright as a subsidiary. IBM has a long history with its own fabs and partners with Albany NanoTech. PsiQuantum's photonic chips come from a partnership with GlobalFoundries, not an owned subsidiary. Quantinuum's manufacturing traces back through Honeywell's aerospace and controls businesses. Owned-and-operated differs from having a good manufacturing partner, and IonQ's claim holds up on that narrow point.",{"type":21,"tag":22,"props":2302,"children":2303},{},[2304,2306,2310,2312,2317],{"type":26,"value":2305},"The claim does not hold up as a physics claim. Owning a foundry produces no new logical qubit and no fidelity gain by itself. Our ",{"type":21,"tag":29,"props":2307,"children":2308},{"href":806},[2309],{"type":26,"value":1706},{"type":26,"value":2311}," and our ",{"type":21,"tag":29,"props":2313,"children":2314},{"href":814},[2315],{"type":26,"value":2316},"SDK and modality comparison",{"type":26,"value":2318}," cover how IonQ stacks up against competitors across every technology, not only this one.",{"type":21,"tag":59,"props":2320,"children":2322},{"id":2321},"why-the-size-of-the-deal-matters",[2323],{"type":26,"value":2324},"Why the size of the deal matters",{"type":21,"tag":22,"props":2326,"children":2327},{},[2328],{"type":26,"value":2329},"$1.8 billion is large next to IonQ's quantum computing revenue history. A large share of the payment was IonQ stock, not cash. SkyWater brings its own profitable semiconductor business, serving customers who have nothing to do with quantum computing. After this deal, IonQ is not a pure quantum computing company anymore. A real share of its revenue now comes from ordinary foundry services. Whether that helps the hardware roadmap or distracts from it is an open question. Future earnings reports will answer that, not this press release.",{"type":21,"tag":59,"props":2331,"children":2333},{"id":2332},"the-domestic-supply-chain-story-holds-up-best",[2334],{"type":26,"value":2335},"The domestic supply chain story holds up best",{"type":21,"tag":22,"props":2337,"children":2338},{},[2339,2341,2347,2348,2353],{"type":26,"value":2340},"SkyWater's federal defense customers and its US-only footprint match a pattern we track elsewhere on this site: the federal push toward domestically controlled, security-vetted quantum and cryptographic infrastructure. See our posts on the ",{"type":21,"tag":29,"props":2342,"children":2344},{"href":2343},"\u002Fblog\u002Fpost-quantum-migration-deadlines",[2345],{"type":26,"value":2346},"accelerated post-quantum migration deadlines",{"type":26,"value":1780},{"type":21,"tag":29,"props":2349,"children":2350},{"href":1274},[2351],{"type":26,"value":2352},"quantum benchmarking executive order",{"type":26,"value":2354},". A quantum hardware vendor with a fully domestic, defense-cleared fabrication and packaging pipeline has a real selling point with government buyers that competitors sourcing through shared or offshore fabs do not have. This part of IonQ's case holds up under scrutiny. It addresses a real, separate requirement: supply chain security, which increasingly gates government contracts regardless of qubit count.",{"type":21,"tag":59,"props":2356,"children":2358},{"id":2357},"what-to-watch-next",[2359],{"type":26,"value":2360},"What to watch next",{"type":21,"tag":22,"props":2362,"children":2363},{},[2364,2366,2371,2373,2378],{"type":26,"value":2365},"A foundry shortens iteration cycles and cuts supply chain risk. It does not improve gate fidelity, connectivity, or logical qubit overhead on its own. Those are the numbers that show whether trapped-ion hardware is progressing. We cover them in ",{"type":21,"tag":29,"props":2367,"children":2368},{"href":751},[2369],{"type":26,"value":2370},"logical qubits and fault tolerance",{"type":26,"value":2372}," and in our look at ",{"type":21,"tag":29,"props":2374,"children":2375},{"href":1141},[2376],{"type":26,"value":2377},"Quantinuum's Helios encoding ratio",{"type":26,"value":2379},". The real test of this acquisition is whether IonQ's next generation of Aria- or Forte-class systems ships faster, or with better specs, than it would have on the old supply chain. That is a multi-year claim, not a same-day one. Until the data arrives, treat \"vertically integrated full-stack quantum platform company\" the way you would treat any vendor headline. Check the structural fact first. SkyWater is now a wholly owned subsidiary. Set the framing aside until results confirm it.",{"title":7,"searchDepth":335,"depth":335,"links":2381},[2382,2383,2384,2385,2386,2387],{"id":2260,"depth":335,"text":2263},{"id":2271,"depth":335,"text":2274},{"id":2292,"depth":335,"text":2295},{"id":2321,"depth":335,"text":2324},{"id":2332,"depth":335,"text":2335},{"id":2357,"depth":335,"text":2360},"content:blog:ionq-skywater-acquisition-explained.md","blog\u002Fionq-skywater-acquisition-explained.md","blog\u002Fionq-skywater-acquisition-explained",{"_path":2392,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":2393,"description":2394,"date":10,"author":11,"tags":2395,"readingTime":474,"body":2396,"_type":715,"_id":2505,"_source":717,"_file":2506,"_stem":2507,"_extension":720},"\u002Fblog\u002Fpsiquantum-darpa-quantum-benchmarking-125-million","DARPA Is Paying PsiQuantum $125 Million to Check Its Own Claims","PsiQuantum signed an expanded $125 million agreement with DARPA's Quantum Benchmarking Initiative, its largest U.S. government contract to date. The money funds outside verification of PsiQuantum's fault-tolerance roadmap, not new research.",[1033,730],{"type":18,"children":2397,"toc":2499},[2398,2419,2424,2430,2435,2440,2446,2451,2463,2469,2474,2479,2483],{"type":21,"tag":22,"props":2399,"children":2400},{},[2401,2403,2409,2410,2417],{"type":26,"value":2402},"On July 22, 2026, PsiQuantum signed an expanded $125 million agreement with the Defense Advanced Research Projects Agency, its largest U.S. government contract to date, according to ",{"type":21,"tag":29,"props":2404,"children":2407},{"href":2405,"rel":2406},"https:\u002F\u002Fthequantuminsider.com\u002F2026\u002F07\u002F22\u002Fpsiquantum-signs-125-million-agreement-with-darpa\u002F",[44],[2408],{"type":26,"value":1499},{"type":26,"value":236},{"type":21,"tag":29,"props":2411,"children":2414},{"href":2412,"rel":2413},"https:\u002F\u002Fwww.psiquantum.com\u002Fnews-import\u002Fpsiquantum-signs-125-million-agreement-with-darpa",[44],[2415],{"type":26,"value":2416},"PsiQuantum's own announcement",{"type":26,"value":2418},". The money sits under DARPA's Quantum Benchmarking Initiative (QBI), a program built around a specific, unglamorous question: is a practically useful quantum computer achievable by 2033, and does any company's roadmap survive an outside party checking it.",{"type":21,"tag":22,"props":2420,"children":2421},{},[2422],{"type":26,"value":2423},"That framing matters more than the dollar figure. This is not research funding for PsiQuantum to build something new. It funds DARPA and independent evaluators testing whether what PsiQuantum has already built and claimed holds up.",{"type":21,"tag":59,"props":2425,"children":2427},{"id":2426},"what-qbi-tests",[2428],{"type":26,"value":2429},"What QBI tests",{"type":21,"tag":22,"props":2431,"children":2432},{},[2433],{"type":26,"value":2434},"DARPA started the Quantum Benchmarking Initiative's predecessor program, Underexplored Systems for Utility-Scale Quantum Computing, in January 2023, when it first selected PsiQuantum for an initial evaluation stage. The program has since progressed in stages: PsiQuantum advanced to a second stage in January 2024, then to Stage C in February 2025, the most advanced tier. In November 2025, DARPA moved 11 companies into an earlier stage of the broader QBI program, and PsiQuantum is one of only two companies reported to have reached Stage C, the tier reserved for the most credible utility-scale claims.",{"type":21,"tag":22,"props":2436,"children":2437},{},[2438],{"type":26,"value":2439},"Getting to Stage C is itself informative. DARPA does not let every company's roadmap through to detailed, funded, adversarial testing. A company that fails an earlier stage is quietly out. Reaching the final tier is a signal that PsiQuantum's claims survived scrutiny that most quantum hardware announcements never face at all.",{"type":21,"tag":59,"props":2441,"children":2443},{"id":2442},"what-the-money-funds",[2444],{"type":26,"value":2445},"What the money funds",{"type":21,"tag":22,"props":2447,"children":2448},{},[2449],{"type":26,"value":2450},"The $125 million follows a smaller $31.8 million agreement from September 2025 that funded on-site testing at PsiQuantum's facilities. This expanded agreement funds testing across PsiQuantum's hardware, software, system architecture, and infrastructure, spanning its Milpitas, California and Chicago, Illinois sites. It is structured as a performance-based award, meaning PsiQuantum gets paid for hitting defined technical milestones DARPA and its evaluators check, not simply for showing up.",{"type":21,"tag":22,"props":2452,"children":2453},{},[2454,2456,2461],{"type":26,"value":2455},"That structure is the point. ",{"type":21,"tag":29,"props":2457,"children":2458},{"href":2181},[2459],{"type":26,"value":2460},"PsiQuantum's photonic, fault-tolerance-first approach",{"type":26,"value":2462}," has never shipped a commercial system, and its timeline for a working fault-tolerant machine has already slipped once, from 2027 to 2029. A company at that stage benefits enormously from being able to say a federal agency is independently verifying its numbers rather than taking PsiQuantum's word for them.",{"type":21,"tag":59,"props":2464,"children":2466},{"id":2465},"why-this-is-a-different-kind-of-evidence-than-a-press-release",[2467],{"type":26,"value":2468},"Why this is a different kind of evidence than a press release",{"type":21,"tag":22,"props":2470,"children":2471},{},[2472],{"type":26,"value":2473},"Most of what this site covers about early-stage quantum hardware comes from company announcements: a qubit count, a fidelity number, a funding round. Those numbers are real, but they are self-reported, and this site treats them accordingly until someone outside the company checks them.",{"type":21,"tag":22,"props":2475,"children":2476},{},[2477],{"type":26,"value":2478},"A DARPA QBI contract is structurally different. The agency's stated purpose is separating hype from reality in quantum computing roadmaps, and it does that by paying independent evaluators to test specific claims rather than accepting a company's own benchmarks. That does not mean PsiQuantum's fault-tolerance timeline will hold. It means the timeline is now subject to a level of outside checking that most competitors in this space do not have applied to their own roadmaps.",{"type":21,"tag":59,"props":2480,"children":2481},{"id":2357},[2482],{"type":26,"value":2360},{"type":21,"tag":22,"props":2484,"children":2485},{},[2486,2488,2492,2493,2497],{"type":26,"value":2487},"QBI's own stated horizon is 2033, so there is no near-term result here to wait for. What is worth tracking is whether PsiQuantum publishes or DARPA discloses any of the specific benchmark results from this testing, rather than only the fact that testing is happening. A contract confirms that evaluation is underway. It does not tell you what the evaluation found. Our ",{"type":21,"tag":29,"props":2489,"children":2490},{"href":806},[2491],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":2494,"children":2495},{"href":862},[2496],{"type":26,"value":1452},{"type":26,"value":2498}," track how PsiQuantum's photonic approach compares with the superconducting, trapped-ion, and neutral-atom platforms other companies are betting on for the same fault-tolerance goal.",{"title":7,"searchDepth":335,"depth":335,"links":2500},[2501,2502,2503,2504],{"id":2426,"depth":335,"text":2429},{"id":2442,"depth":335,"text":2445},{"id":2465,"depth":335,"text":2468},{"id":2357,"depth":335,"text":2360},"content:blog:psiquantum-darpa-quantum-benchmarking-125-million.md","blog\u002Fpsiquantum-darpa-quantum-benchmarking-125-million.md","blog\u002Fpsiquantum-darpa-quantum-benchmarking-125-million",{"_path":2509,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":2510,"description":2511,"date":10,"author":11,"tags":2512,"readingTime":453,"body":2513,"_type":715,"_id":2607,"_source":717,"_file":2608,"_stem":2609,"_extension":720},"\u002Fblog\u002Fquantinuum-bmw-multi-year-partnership","Quantinuum and BMW Turned a Five-Year Research Project Into a Roadmap Commitment","Quantinuum and BMW Group expanded their quantum chemistry collaboration, running since 2021, into a multi-year partnership that gives BMW access to Quantinuum's Helios, Sol, and Apollo hardware generations through 2029. Here is what five years of actual results looks like next to a typical pilot announcement.",[1033,730],{"type":18,"children":2514,"toc":2602},[2515,2536,2542,2547,2552,2558,2570,2575,2581,2586],{"type":21,"tag":22,"props":2516,"children":2517},{},[2518,2520,2527,2528,2534],{"type":26,"value":2519},"On May 5, 2026, Quantinuum and BMW Group announced they were expanding their quantum computing collaboration into a formal multi-year partnership, according to ",{"type":21,"tag":29,"props":2521,"children":2524},{"href":2522,"rel":2523},"https:\u002F\u002Fwww.quantinuum.com\u002Fpress-releases\u002Fquantinuum-and-bmw-group-expand-landmark-quantum-computing-collaboration-with-new-multi-year-partnership",[44],[2525],{"type":26,"value":2526},"Quantinuum's press release",{"type":26,"value":236},{"type":21,"tag":29,"props":2529,"children":2532},{"href":2530,"rel":2531},"https:\u002F\u002Fthequantuminsider.com\u002F2026\u002F05\u002F05\u002Fquantinuum-and-bmw-group-expand-quantum-computing-collaboration-with-new-multi-year-partnership\u002F",[44],[2533],{"type":26,"value":1499},{"type":26,"value":2535},". This is not a new story breaking this week. It is worth covering anyway, because most quantum-for-industry partnerships announced with this much fanfare are pilots with no track record behind them, and this one has run since 2021.",{"type":21,"tag":59,"props":2537,"children":2539},{"id":2538},"five-years-is-the-actual-story-here",[2540],{"type":26,"value":2541},"Five years is the actual story here",{"type":21,"tag":22,"props":2543,"children":2544},{},[2545],{"type":26,"value":2546},"Quantinuum and BMW have collaborated on industrial chemistry problems since 2021, moving from early algorithm development to simulating real molecular systems relevant to next-generation mobility. The specific research area is electrochemistry for fuel cells and batteries, including modeling the oxygen reduction reaction at platinum catalysts, a problem that determines fuel cell efficiency and where classical simulation methods struggle with the electron correlation effects involved.",{"type":21,"tag":22,"props":2548,"children":2549},{},[2550],{"type":26,"value":2551},"That history matters more than the announcement itself. Most quantum industry partnerships get covered at the moment they are signed, before there is any way to know whether the collaboration produces results or quietly fades. A five-year run with published progress from early algorithm work to real molecular simulation is a track record, not a press release promise.",{"type":21,"tag":59,"props":2553,"children":2555},{"id":2554},"what-bmw-gets",[2556],{"type":26,"value":2557},"What BMW gets",{"type":21,"tag":22,"props":2559,"children":2560},{},[2561,2563,2568],{"type":26,"value":2562},"The expanded partnership gives BMW access to successive generations of Quantinuum's ",{"type":21,"tag":29,"props":2564,"children":2565},{"href":1141},[2566],{"type":26,"value":2567},"trapped-ion",{"type":26,"value":2569}," hardware on a defined timeline: Helios (the current 96-qubit system, available now as hardware-as-a-service), Sol (planned for 2027 as Quantinuum's first commercially available system built on a two-dimensional qubit grid), and Apollo (targeted for 2029 as a fully fault-tolerant system capable of executing millions of gates).",{"type":21,"tag":22,"props":2571,"children":2572},{},[2573],{"type":26,"value":2574},"Quantinuum CEO Rajeeb Hazra framed the deal around \"driving commercial adoption of quantum computing through close collaboration with industry leaders on high-impact applications.\" BMW's Vice President of New Technologies, Martin Tietze, said the companies \"translate advances in quantum hardware into real-world applications.\" Both statements are the kind of framing a joint press release always uses. The part that is independently checkable is the hardware access itself: BMW is committed to Quantinuum's roadmap through at least 2029, which is an unusually long horizon for a customer relationship in a field where most vendor roadmaps get revised within a year or two.",{"type":21,"tag":59,"props":2576,"children":2578},{"id":2577},"why-this-is-a-useful-comparison-point",[2579],{"type":26,"value":2580},"Why this is a useful comparison point",{"type":21,"tag":22,"props":2582,"children":2583},{},[2584],{"type":26,"value":2585},"Announcements about quantum computing partnerships with automakers, banks, and pharmaceutical companies are common, and most of them describe an exploratory pilot with no committed timeline and no disclosed technical result. This partnership is a useful yardstick against those: a five-year history of published, incremental research progress, a named scientific problem (the oxygen reduction reaction) rather than a vague reference to \"materials science,\" and a hardware access commitment tied to specific, dated system generations rather than an open-ended promise to keep exploring.",{"type":21,"tag":22,"props":2587,"children":2588},{},[2589,2591,2595,2596,2600],{"type":26,"value":2590},"None of that guarantees the electrochemistry results translate into a better BMW battery or fuel cell. It does mean this collaboration has a longer, more specific paper trail than the average industry quantum pilot, and it is worth checking back against as Sol and Apollo ship, since the value of a roadmap commitment depends entirely on whether the hardware behind it arrives on schedule. Our ",{"type":21,"tag":29,"props":2592,"children":2593},{"href":806},[2594],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":2597,"children":2598},{"href":862},[2599],{"type":26,"value":1452},{"type":26,"value":2601}," track how Quantinuum's trapped-ion roadmap compares with the superconducting, neutral-atom, and photonic platforms other vendors are building toward the same industrial-chemistry use cases.",{"title":7,"searchDepth":335,"depth":335,"links":2603},[2604,2605,2606],{"id":2538,"depth":335,"text":2541},{"id":2554,"depth":335,"text":2557},{"id":2577,"depth":335,"text":2580},"content:blog:quantinuum-bmw-multi-year-partnership.md","blog\u002Fquantinuum-bmw-multi-year-partnership.md","blog\u002Fquantinuum-bmw-multi-year-partnership",{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"author":11,"tags":2611,"readingTime":16,"body":2612,"_type":715,"_id":716,"_source":717,"_file":718,"_stem":719,"_extension":720},[13,14,15],{"type":18,"children":2613,"toc":3139},[2614,2623,2637,2641,2668,2672,2676,2684,2703,2729,2733,2737,2741,2754,2776,2780,2799,2825,2994,3072,3086,3094,3098,3107,3135],{"type":21,"tag":22,"props":2615,"children":2616},{},[2617,2618,2622],{"type":26,"value":27},{"type":21,"tag":29,"props":2619,"children":2620},{"href":31},[2621],{"type":26,"value":34},{"type":26,"value":36},{"type":21,"tag":22,"props":2624,"children":2625},{},[2626,2631,2632,2636],{"type":21,"tag":29,"props":2627,"children":2629},{"href":42,"rel":2628},[44],[2630],{"type":26,"value":47},{"type":26,"value":49},{"type":21,"tag":51,"props":2633,"children":2634},{},[2635],{"type":26,"value":55},{"type":26,"value":57},{"type":21,"tag":59,"props":2638,"children":2639},{"id":61},[2640],{"type":26,"value":64},{"type":21,"tag":22,"props":2642,"children":2643},{},[2644,2645,2649,2650,2655,2656,2661,2662,2667],{"type":26,"value":69},{"type":21,"tag":71,"props":2646,"children":2647},{},[2648],{"type":26,"value":75},{"type":26,"value":77},{"type":21,"tag":79,"props":2651,"children":2653},{"className":2652},[],[2654],{"type":26,"value":84},{"type":26,"value":86},{"type":21,"tag":79,"props":2657,"children":2659},{"className":2658},[],[2660],{"type":26,"value":92},{"type":26,"value":94},{"type":21,"tag":79,"props":2663,"children":2665},{"className":2664},[],[2666],{"type":26,"value":100},{"type":26,"value":102},{"type":21,"tag":59,"props":2669,"children":2670},{"id":105},[2671],{"type":26,"value":108},{"type":21,"tag":22,"props":2673,"children":2674},{},[2675],{"type":26,"value":113},{"type":21,"tag":22,"props":2677,"children":2678},{},[2679,2683],{"type":21,"tag":71,"props":2680,"children":2681},{},[2682],{"type":26,"value":121},{"type":26,"value":123},{"type":21,"tag":22,"props":2685,"children":2686},{},[2687,2691,2692,2697,2698,2702],{"type":21,"tag":71,"props":2688,"children":2689},{},[2690],{"type":26,"value":131},{"type":26,"value":133},{"type":21,"tag":79,"props":2693,"children":2695},{"className":2694},[],[2696],{"type":26,"value":139},{"type":26,"value":141},{"type":21,"tag":29,"props":2699,"children":2700},{"href":144},[2701],{"type":26,"value":14},{"type":26,"value":148},{"type":21,"tag":22,"props":2704,"children":2705},{},[2706,2710,2711,2716,2717,2722,2723,2728],{"type":21,"tag":71,"props":2707,"children":2708},{},[2709],{"type":26,"value":156},{"type":26,"value":158},{"type":21,"tag":79,"props":2712,"children":2714},{"className":2713},[],[2715],{"type":26,"value":164},{"type":26,"value":166},{"type":21,"tag":79,"props":2718,"children":2720},{"className":2719},[],[2721],{"type":26,"value":172},{"type":26,"value":174},{"type":21,"tag":79,"props":2724,"children":2726},{"className":2725},[],[2727],{"type":26,"value":180},{"type":26,"value":182},{"type":21,"tag":22,"props":2730,"children":2731},{},[2732],{"type":26,"value":187},{"type":21,"tag":59,"props":2734,"children":2735},{"id":190},[2736],{"type":26,"value":193},{"type":21,"tag":22,"props":2738,"children":2739},{},[2740],{"type":26,"value":198},{"type":21,"tag":200,"props":2742,"children":2743},{},[2744],{"type":21,"tag":22,"props":2745,"children":2746},{},[2747,2748,2753],{"type":26,"value":207},{"type":21,"tag":79,"props":2749,"children":2751},{"className":2750},[],[2752],{"type":26,"value":213},{"type":26,"value":215},{"type":21,"tag":22,"props":2755,"children":2756},{},[2757,2758,2763,2764,2769,2770,2775],{"type":26,"value":220},{"type":21,"tag":79,"props":2759,"children":2761},{"className":2760},[],[2762],{"type":26,"value":226},{"type":26,"value":228},{"type":21,"tag":79,"props":2765,"children":2767},{"className":2766},[],[2768],{"type":26,"value":234},{"type":26,"value":236},{"type":21,"tag":79,"props":2771,"children":2773},{"className":2772},[],[2774],{"type":26,"value":242},{"type":26,"value":244},{"type":21,"tag":59,"props":2777,"children":2778},{"id":247},[2779],{"type":26,"value":250},{"type":21,"tag":22,"props":2781,"children":2782},{},[2783,2787,2788,2792,2793,2798],{"type":21,"tag":71,"props":2784,"children":2785},{},[2786],{"type":26,"value":258},{"type":26,"value":260},{"type":21,"tag":29,"props":2789,"children":2790},{"href":263},[2791],{"type":26,"value":266},{"type":26,"value":268},{"type":21,"tag":79,"props":2794,"children":2796},{"className":2795},[],[2797],{"type":26,"value":274},{"type":26,"value":276},{"type":21,"tag":22,"props":2800,"children":2801},{},[2802,2806,2807,2812,2813,2818,2819,2824],{"type":21,"tag":71,"props":2803,"children":2804},{},[2805],{"type":26,"value":284},{"type":26,"value":286},{"type":21,"tag":79,"props":2808,"children":2810},{"className":2809},[],[2811],{"type":26,"value":292},{"type":26,"value":294},{"type":21,"tag":79,"props":2814,"children":2816},{"className":2815},[],[2817],{"type":26,"value":300},{"type":26,"value":302},{"type":21,"tag":79,"props":2820,"children":2822},{"className":2821},[],[2823],{"type":26,"value":308},{"type":26,"value":310},{"type":21,"tag":312,"props":2826,"children":2827},{"className":314,"code":315,"language":316,"meta":7,"style":7},[2828],{"type":21,"tag":79,"props":2829,"children":2830},{"__ignoreMap":7},[2831,2838,2845,2896,2915,2934,2953,2972,2987],{"type":21,"tag":322,"props":2832,"children":2833},{"class":324,"line":325},[2834],{"type":21,"tag":322,"props":2835,"children":2836},{"style":329},[2837],{"type":26,"value":332},{"type":21,"tag":322,"props":2839,"children":2840},{"class":324,"line":335},[2841],{"type":21,"tag":322,"props":2842,"children":2843},{"style":329},[2844],{"type":26,"value":341},{"type":21,"tag":322,"props":2846,"children":2847},{"class":324,"line":344},[2848,2852,2856,2860,2864,2868,2872,2876,2880,2884,2888,2892],{"type":21,"tag":322,"props":2849,"children":2850},{"style":348},[2851],{"type":26,"value":351},{"type":21,"tag":322,"props":2853,"children":2854},{"style":329},[2855],{"type":26,"value":356},{"type":21,"tag":322,"props":2857,"children":2858},{"style":359},[2859],{"type":26,"value":362},{"type":21,"tag":322,"props":2861,"children":2862},{"style":329},[2863],{"type":26,"value":367},{"type":21,"tag":322,"props":2865,"children":2866},{"style":359},[2867],{"type":26,"value":372},{"type":21,"tag":322,"props":2869,"children":2870},{"style":329},[2871],{"type":26,"value":367},{"type":21,"tag":322,"props":2873,"children":2874},{"style":359},[2875],{"type":26,"value":381},{"type":21,"tag":322,"props":2877,"children":2878},{"style":329},[2879],{"type":26,"value":367},{"type":21,"tag":322,"props":2881,"children":2882},{"style":359},[2883],{"type":26,"value":390},{"type":21,"tag":322,"props":2885,"children":2886},{"style":329},[2887],{"type":26,"value":367},{"type":21,"tag":322,"props":2889,"children":2890},{"style":359},[2891],{"type":26,"value":399},{"type":21,"tag":322,"props":2893,"children":2894},{"style":329},[2895],{"type":26,"value":404},{"type":21,"tag":322,"props":2897,"children":2898},{"class":324,"line":407},[2899,2903,2907,2911],{"type":21,"tag":322,"props":2900,"children":2901},{"style":348},[2902],{"type":26,"value":413},{"type":21,"tag":322,"props":2904,"children":2905},{"style":329},[2906],{"type":26,"value":418},{"type":21,"tag":322,"props":2908,"children":2909},{"style":348},[2910],{"type":26,"value":423},{"type":21,"tag":322,"props":2912,"children":2913},{"style":329},[2914],{"type":26,"value":428},{"type":21,"tag":322,"props":2916,"children":2917},{"class":324,"line":431},[2918,2922,2926,2930],{"type":21,"tag":322,"props":2919,"children":2920},{"style":348},[2921],{"type":26,"value":437},{"type":21,"tag":322,"props":2923,"children":2924},{"style":329},[2925],{"type":26,"value":418},{"type":21,"tag":322,"props":2927,"children":2928},{"style":348},[2929],{"type":26,"value":446},{"type":21,"tag":322,"props":2931,"children":2932},{"style":329},[2933],{"type":26,"value":428},{"type":21,"tag":322,"props":2935,"children":2936},{"class":324,"line":453},[2937,2941,2945,2949],{"type":21,"tag":322,"props":2938,"children":2939},{"style":348},[2940],{"type":26,"value":459},{"type":21,"tag":322,"props":2942,"children":2943},{"style":329},[2944],{"type":26,"value":418},{"type":21,"tag":322,"props":2946,"children":2947},{"style":348},[2948],{"type":26,"value":423},{"type":21,"tag":322,"props":2950,"children":2951},{"style":329},[2952],{"type":26,"value":428},{"type":21,"tag":322,"props":2954,"children":2955},{"class":324,"line":474},[2956,2960,2964,2968],{"type":21,"tag":322,"props":2957,"children":2958},{"style":348},[2959],{"type":26,"value":480},{"type":21,"tag":322,"props":2961,"children":2962},{"style":329},[2963],{"type":26,"value":418},{"type":21,"tag":322,"props":2965,"children":2966},{"style":348},[2967],{"type":26,"value":489},{"type":21,"tag":322,"props":2969,"children":2970},{"style":329},[2971],{"type":26,"value":428},{"type":21,"tag":322,"props":2973,"children":2974},{"class":324,"line":496},[2975,2979,2983],{"type":21,"tag":322,"props":2976,"children":2977},{"style":348},[2978],{"type":26,"value":502},{"type":21,"tag":322,"props":2980,"children":2981},{"style":329},[2982],{"type":26,"value":418},{"type":21,"tag":322,"props":2984,"children":2985},{"style":359},[2986],{"type":26,"value":511},{"type":21,"tag":322,"props":2988,"children":2989},{"class":324,"line":16},[2990],{"type":21,"tag":322,"props":2991,"children":2992},{"style":329},[2993],{"type":26,"value":519},{"type":21,"tag":22,"props":2995,"children":2996},{},[2997,3007,3008,3013,3014,3019,3020,3024,3025,3030,3031,3036,3037,3042,3043,3048,3049,3053,3054,3059,3060,3065,3066,3071],{"type":21,"tag":71,"props":2998,"children":2999},{},[3000,3001,3006],{"type":26,"value":527},{"type":21,"tag":79,"props":3002,"children":3004},{"className":3003},[],[3005],{"type":26,"value":213},{"type":26,"value":534},{"type":26,"value":536},{"type":21,"tag":79,"props":3009,"children":3011},{"className":3010},[],[3012],{"type":26,"value":542},{"type":26,"value":544},{"type":21,"tag":79,"props":3015,"children":3017},{"className":3016},[],[3018],{"type":26,"value":550},{"type":26,"value":552},{"type":21,"tag":29,"props":3021,"children":3022},{"href":555},[3023],{"type":26,"value":558},{"type":26,"value":560},{"type":21,"tag":79,"props":3026,"children":3028},{"className":3027},[],[3029],{"type":26,"value":213},{"type":26,"value":567},{"type":21,"tag":79,"props":3032,"children":3034},{"className":3033},[],[3035],{"type":26,"value":573},{"type":26,"value":236},{"type":21,"tag":79,"props":3038,"children":3040},{"className":3039},[],[3041],{"type":26,"value":242},{"type":26,"value":581},{"type":21,"tag":79,"props":3044,"children":3046},{"className":3045},[],[3047],{"type":26,"value":213},{"type":26,"value":588},{"type":21,"tag":51,"props":3050,"children":3051},{},[3052],{"type":26,"value":593},{"type":26,"value":595},{"type":21,"tag":79,"props":3055,"children":3057},{"className":3056},[],[3058],{"type":26,"value":601},{"type":26,"value":603},{"type":21,"tag":79,"props":3061,"children":3063},{"className":3062},[],[3064],{"type":26,"value":609},{"type":26,"value":611},{"type":21,"tag":79,"props":3067,"children":3069},{"className":3068},[],[3070],{"type":26,"value":617},{"type":26,"value":619},{"type":21,"tag":22,"props":3073,"children":3074},{},[3075,3079,3080,3085],{"type":21,"tag":71,"props":3076,"children":3077},{},[3078],{"type":26,"value":627},{"type":26,"value":629},{"type":21,"tag":79,"props":3081,"children":3083},{"className":3082},[],[3084],{"type":26,"value":308},{"type":26,"value":636},{"type":21,"tag":22,"props":3087,"children":3088},{},[3089,3093],{"type":21,"tag":71,"props":3090,"children":3091},{},[3092],{"type":26,"value":644},{"type":26,"value":646},{"type":21,"tag":59,"props":3095,"children":3096},{"id":649},[3097],{"type":26,"value":652},{"type":21,"tag":22,"props":3099,"children":3100},{},[3101,3102,3106],{"type":26,"value":657},{"type":21,"tag":29,"props":3103,"children":3104},{"href":31},[3105],{"type":26,"value":662},{"type":26,"value":664},{"type":21,"tag":22,"props":3108,"children":3109},{},[3110,3111,3116,3117,3122,3123,3128,3129,3134],{"type":26,"value":669},{"type":21,"tag":29,"props":3112,"children":3114},{"href":42,"rel":3113},[44],[3115],{"type":26,"value":675},{"type":26,"value":677},{"type":21,"tag":79,"props":3118,"children":3120},{"className":3119},[],[3121],{"type":26,"value":683},{"type":26,"value":685},{"type":21,"tag":79,"props":3124,"children":3126},{"className":3125},[],[3127],{"type":26,"value":691},{"type":26,"value":693},{"type":21,"tag":79,"props":3130,"children":3132},{"className":3131},[],[3133],{"type":26,"value":699},{"type":26,"value":701},{"type":21,"tag":703,"props":3136,"children":3137},{},[3138],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":3140},[3141,3142,3143,3144,3145],{"id":61,"depth":335,"text":64},{"id":105,"depth":335,"text":108},{"id":190,"depth":335,"text":193},{"id":247,"depth":335,"text":250},{"id":649,"depth":335,"text":652},{"_path":3147,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":3148,"description":3149,"date":10,"author":11,"tags":3150,"readingTime":474,"body":3151,"_type":715,"_id":3295,"_source":717,"_file":3296,"_stem":3297,"_extension":720},"\u002Fblog\u002Fsaxon-q-room-temperature-diamond-quantum-computer","SAXON Q Is Selling a Quantum Computer That Skips the Cryostat Entirely","SAXON Q, a Leipzig University spinout, launched the SXQ128 and SXQ512, diamond-based nitrogen-vacancy quantum computers that run at room temperature with no cryogenic cooling. Here is how the qubits work and what the claimed numbers mean.",[730,1033],{"type":18,"children":3152,"toc":3289},[3153,3174,3179,3185,3199,3204,3210,3222,3227,3233,3245,3250,3256,3268,3284],{"type":21,"tag":22,"props":3154,"children":3155},{},[3156,3158,3164,3165,3172],{"type":26,"value":3157},"On July 21, 2026, SAXON Q announced commercial availability of the SXQ128 and SXQ512, diamond-based quantum computers built around nitrogen-vacancy (NV) centers, according to ",{"type":21,"tag":29,"props":3159,"children":3162},{"href":3160,"rel":3161},"https:\u002F\u002Fthequantuminsider.com\u002F2026\u002F07\u002F21\u002Fsaxon-q-diamond-nv-center-quantum-computers\u002F",[44],[3163],{"type":26,"value":1499},{"type":26,"value":236},{"type":21,"tag":29,"props":3166,"children":3169},{"href":3167,"rel":3168},"https:\u002F\u002Fwww.hpcwire.com\u002Foff-the-wire\u002Fsaxon-q-brings-room-temperature-diamond-quantum-computing-to-commercial-market\u002F",[44],[3170],{"type":26,"value":3171},"HPCwire",{"type":26,"value":3173},". The headline claim is the one worth checking first: both systems run at room temperature, with no cryostat, no vacuum chamber, and no dilution refrigerator, plugged into an ordinary electrical outlet in a standard server rack.",{"type":21,"tag":22,"props":3175,"children":3176},{},[3177],{"type":26,"value":3178},"That single fact separates NV-center qubits from every hardware modality this site usually covers. Superconducting qubits and trapped ions both need cooling to near absolute zero, and most of the engineering difficulty in scaling those platforms traces back to that requirement. A qubit that works at room temperature sidesteps an entire category of problems, if the claim holds up.",{"type":21,"tag":59,"props":3180,"children":3182},{"id":3181},"what-saxon-q-built",[3183],{"type":26,"value":3184},"What SAXON Q built",{"type":21,"tag":22,"props":3186,"children":3187},{},[3188,3190,3197],{"type":26,"value":3189},"SAXON Q is a 2021 spinout from Leipzig University, founded by physicists who spent decades researching NV centers before commercializing the work. Marius Grundmann and Frank Schlichting hold the CEO title jointly, according to ",{"type":21,"tag":29,"props":3191,"children":3194},{"href":3192,"rel":3193},"https:\u002F\u002Fwww.saxonq.com\u002Fen\u002Funternehmen\u002F",[44],[3195],{"type":26,"value":3196},"the company's own site",{"type":26,"value":3198},", alongside co-founders Jan Meijer (CTO) and Bernd Burchard (CIPO).",{"type":21,"tag":22,"props":3200,"children":3201},{},[3202],{"type":26,"value":3203},"The SXQ128 packs 128 qubits, arranged as multiple cores of eight fully entangled qubits each. The SXQ512 scales that to 512 qubits across cores of 16. Both are available to order now, with SXQ128 deliveries beginning within three months and SXQ512 shipments starting in the second quarter of 2027.",{"type":21,"tag":59,"props":3205,"children":3207},{"id":3206},"what-an-nv-center-is",[3208],{"type":26,"value":3209},"What an NV center is",{"type":21,"tag":22,"props":3211,"children":3212},{},[3213,3215,3220],{"type":26,"value":3214},"A nitrogen-vacancy center is a specific defect in a diamond's carbon lattice: a nitrogen atom sits next to an empty spot where a carbon atom should be. That defect traps a single electron whose spin state is initialized, manipulated, and read out with lasers and microwaves, and it behaves as a ",{"type":21,"tag":29,"props":3216,"children":3217},{"href":1794},[3218],{"type":26,"value":3219},"qubit",{"type":26,"value":3221}," at room temperature because diamond's rigid lattice isolates the trapped electron from the thermal noise that would otherwise destroy its quantum state.",{"type":21,"tag":22,"props":3223,"children":3224},{},[3225],{"type":26,"value":3226},"The company reports a fidelity of up to 99.92% and describes creating these defects through a proprietary sulfur co-implantation process during ion beam implantation, claiming a conversion yield above 85% (the share of implanted atoms that become working qubits) against a reported 1 to 10% for prior implantation methods. That yield number is the more interesting engineering claim here, since NV centers themselves are decades-old physics. The obstacle was always building enough of them reliably in one piece of diamond, not discovering that they exist.",{"type":21,"tag":59,"props":3228,"children":3230},{"id":3229},"what-to-treat-as-unverified",[3231],{"type":26,"value":3232},"What to treat as unverified",{"type":21,"tag":22,"props":3234,"children":3235},{},[3236,3238,3243],{"type":26,"value":3237},"Every number above comes from SAXON Q's own announcement. None of it has been independently reproduced or published in a peer-reviewed venue the way HRL's recent ",{"type":21,"tag":29,"props":3239,"children":3240},{"href":1617},[3241],{"type":26,"value":3242},"self-operating silicon processor",{"type":26,"value":3244}," was. A fidelity figure is only useful next to a clearly stated definition of which operation it measures, and a yield percentage is only useful once someone outside the company that reports it gets to check the diamond.",{"type":21,"tag":22,"props":3246,"children":3247},{},[3248],{"type":26,"value":3249},"The one data point that is not only a press release is a deployment: Fraunhofer IWU installed a SAXON Q system in mid-2025 for industrial optimization work in material processing and robotics, and the company says it has run continuously at room temperature since. A working deployment at a respected applied-research institute is a stronger signal than a spec sheet, though it still says nothing about how the same architecture performs at 128 or 512 qubits.",{"type":21,"tag":59,"props":3251,"children":3253},{"id":3252},"where-room-temperature-qubits-fit-against-everything-else",[3254],{"type":26,"value":3255},"Where room-temperature qubits fit against everything else",{"type":21,"tag":22,"props":3257,"children":3258},{},[3259,3261,3266],{"type":26,"value":3260},"Room-temperature operation does not automatically mean SAXON Q's qubits outperform cryogenic ones on the metrics that decide whether a quantum computer is useful: gate fidelity under load, coherence time during a real circuit, and how the system behaves as qubit count grows. Trapped ions and superconducting qubits have a multi-year head start on demonstrating those things at scale, with ",{"type":21,"tag":29,"props":3262,"children":3263},{"href":1274},[3264],{"type":26,"value":3265},"independently scrutinized results",{"type":26,"value":3267}," that NV-center systems do not yet have.",{"type":21,"tag":22,"props":3269,"children":3270},{},[3271,3273,3277,3278,3282],{"type":26,"value":3272},"What room temperature does buy, if the claims hold, is dramatically simpler infrastructure. No cryostat means no multi-week cooldown cycle, no liquid helium supply chain, and a system that fits in a standard rack instead of a dedicated lab. For applications like industrial optimization at a manufacturing site, that operational simplicity sometimes matters more than an extra order of magnitude of fidelity. Our ",{"type":21,"tag":29,"props":3274,"children":3275},{"href":806},[3276],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":3279,"children":3280},{"href":862},[3281],{"type":26,"value":1452},{"type":26,"value":3283}," track how each modality trades those priorities differently.",{"type":21,"tag":22,"props":3285,"children":3286},{},[3287],{"type":26,"value":3288},"The honest read on SAXON Q today: a real, shipping product from a team with a genuine research background in the underlying physics, with commercial claims that have not yet been checked by anyone outside the company. That is a normal place for a new hardware vendor to be. It is not yet evidence that diamond NV centers compete with the platforms that have already survived a decade of outside scrutiny.",{"title":7,"searchDepth":335,"depth":335,"links":3290},[3291,3292,3293,3294],{"id":3181,"depth":335,"text":3184},{"id":3206,"depth":335,"text":3209},{"id":3229,"depth":335,"text":3232},{"id":3252,"depth":335,"text":3255},"content:blog:saxon-q-room-temperature-diamond-quantum-computer.md","blog\u002Fsaxon-q-room-temperature-diamond-quantum-computer.md","blog\u002Fsaxon-q-room-temperature-diamond-quantum-computer",{"_path":1691,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":3299,"description":3300,"date":10,"author":11,"tags":3301,"readingTime":453,"body":3303,"_type":715,"_id":3385,"_source":717,"_file":3386,"_stem":3387,"_extension":720},"Warwick's New Chip Concept Aims to Wire Together a Million Qubits","Warwick and NRC Canada proposed Quantum Phononic Links, a way to carry quantum information across an entire chip using engineered vibrations instead of only connecting neighboring qubits.",[730,1033,3302],"Fundamentals",{"type":18,"children":3304,"toc":3380},[3305,3310,3316,3321,3340,3346,3358,3363,3369],{"type":21,"tag":22,"props":3306,"children":3307},{},[3308],{"type":26,"value":3309},"Today's leading quantum chips mostly connect a qubit to its immediate neighbors and nothing else. A useful large-scale quantum computer needs coordination across millions of qubits spread over an entire chip, not only clusters sitting next to each other. Researchers at the University of Warwick and NRC Canada published a proposed answer to that gap on July 29, 2026, in APL Quantum: a concept they call Quantum Phononic Links.",{"type":21,"tag":59,"props":3311,"children":3313},{"id":3312},"what-a-phononic-link-is",[3314],{"type":26,"value":3315},"What a phononic link is",{"type":21,"tag":22,"props":3317,"children":3318},{},[3319],{"type":26,"value":3320},"The idea uses phonons, sound-like vibrations traveling through a solid material, to carry quantum information between qubits that sit far apart on the same chip. The material behind it, compressively strained germanium on silicon, was developed at Warwick using advanced epitaxial growth techniques. Engineered correctly, the vibrations carry information whether the qubits involved sit side by side or are separated across a full semiconductor wafer up to 300 millimeters across.",{"type":21,"tag":22,"props":3322,"children":3323},{},[3324,3326,3331,3333,3338],{"type":26,"value":3325},"That distance matters. Nearest-neighbor-only connectivity is exactly the constraint that shapes which ",{"type":21,"tag":29,"props":3327,"children":3328},{"href":1526},[3329],{"type":26,"value":3330},"error-correcting codes",{"type":26,"value":3332}," a chip even attempts. Our ",{"type":21,"tag":29,"props":3334,"children":3335},{"href":1141},[3336],{"type":26,"value":3337},"companion piece on Quantinuum's Helios",{"type":26,"value":3339}," covers why trapped-ion hardware's all-to-all connectivity opens up error-correction codes that a nearest-neighbor superconducting chip cannot use efficiently. A phononic link aimed at chip-scale, non-local connectivity is targeting the same structural limitation from the silicon side.",{"type":21,"tag":59,"props":3341,"children":3343},{"id":3342},"read-the-maturity-level-accurately",[3344],{"type":26,"value":3345},"Read the maturity level accurately",{"type":21,"tag":22,"props":3347,"children":3348},{},[3349,3351,3356],{"type":26,"value":3350},"This is a concept and materials result, not an operating multi-qubit device. The paper proposes a mechanism and demonstrates the underlying material properties. It does not yet show a working chip moving quantum information between distant qubits using this method. That puts it in a different category of evidence than ",{"type":21,"tag":29,"props":3352,"children":3353},{"href":1617},[3354],{"type":26,"value":3355},"HRL's self-operating silicon processor",{"type":26,"value":3357},", published the same week, which is a working 18-qubit device with a control chip already running inside its cryostat.",{"type":21,"tag":22,"props":3359,"children":3360},{},[3361],{"type":26,"value":3362},"Both results matter, and they matter for different reasons. HRL solved a real engineering problem for a small, existing device. Warwick proposed a mechanism aimed at a target two or three orders of magnitude larger, with the harder work of building and testing an actual chip still ahead of it. Neither is more important than the other. They sit at different points on the path from proposal to product, and coverage that treats them as equally proven misses that distinction.",{"type":21,"tag":59,"props":3364,"children":3366},{"id":3365},"why-the-target-size-is-the-real-headline",[3367],{"type":26,"value":3368},"Why the target size is the real headline",{"type":21,"tag":22,"props":3370,"children":3371},{},[3372,3374,3378],{"type":26,"value":3373},"A million qubits is well beyond any current device, superconducting, trapped-ion, or silicon. Getting there needs progress on several fronts at once: qubit quality, error correction overhead, control electronics, and exactly the qubit-to-qubit communication problem this proposal addresses. No single result solves all of it. What is worth tracking is whether Warwick's group, or others building on the same cs-GoS material, demonstrates the mechanism working on an actual multi-qubit chip. That demonstration, not the concept paper, will be the point where this result earns a place next to HRL's on the list of things that shipped. See our ",{"type":21,"tag":29,"props":3375,"children":3376},{"href":806},[3377],{"type":26,"value":1706},{"type":26,"value":3379}," for how silicon-based approaches compare to the rest of the field.",{"title":7,"searchDepth":335,"depth":335,"links":3381},[3382,3383,3384],{"id":3312,"depth":335,"text":3315},{"id":3342,"depth":335,"text":3345},{"id":3365,"depth":335,"text":3368},"content:blog:warwick-quantum-phononic-links.md","blog\u002Fwarwick-quantum-phononic-links.md","blog\u002Fwarwick-quantum-phononic-links",{"_path":3389,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":3390,"description":3391,"date":10,"author":11,"tags":3392,"readingTime":474,"body":3393,"_type":715,"_id":3493,"_source":717,"_file":3494,"_stem":3495,"_extension":720},"\u002Fblog\u002Fzuriq-2d-trapped-ion-seed-funding","ZuriQ Wants to Fix Trapped-Ion Scaling by Going 2D Instead of 1D","ZuriQ, an ETH Zurich spinout, raised $25.5 million to scale a two-dimensional trapped-ion architecture built on Penning micro-traps, a departure from the linear ion chains IonQ and Quantinuum use. Here is what the nine-ion demonstrator shows.",[730,1033,731],{"type":18,"children":3394,"toc":3488},[3395,3415,3426,3432,3444,3449,3455,3460,3465,3471,3476],{"type":21,"tag":22,"props":3396,"children":3397},{},[3398,3400,3406,3407,3413],{"type":26,"value":3399},"On July 28, 2026, ZuriQ announced a $25.5 million seed round to scale a trapped-ion architecture that departs from how every other trapped-ion company builds its qubits, according to ",{"type":21,"tag":29,"props":3401,"children":3404},{"href":3402,"rel":3403},"https:\u002F\u002Fthequantuminsider.com\u002F2026\u002F07\u002F28\u002Fzuriq-raises-25-5-million-2d-trapped-ion-quantum-processors\u002F",[44],[3405],{"type":26,"value":1499},{"type":26,"value":236},{"type":21,"tag":29,"props":3408,"children":3411},{"href":3409,"rel":3410},"https:\u002F\u002Fsiliconangle.com\u002F2026\u002F07\u002F28\u002Fquantum-computing-startup-zuriq-gets-25-5m-scale-2d-trapped-ion-processor-design\u002F",[44],[3412],{"type":26,"value":1922},{"type":26,"value":3414},". The round was led by Quantonation, with Forward.one, Extantia, and Firgun Ventures joining alongside existing investors, following a $4.2 million pre-seed round in 2025.",{"type":21,"tag":22,"props":3416,"children":3417},{},[3418,3420,3424],{"type":26,"value":3419},"ZuriQ is a spinout from ETH Zurich, led by CEO Pavel Hrmo, who spent more than a decade in trapped-ion research before founding the company. Its pitch is narrow and specific: the standard way of building a ",{"type":21,"tag":29,"props":3421,"children":3422},{"href":806},[3423],{"type":26,"value":2567},{"type":26,"value":3425}," quantum computer, a linear chain of ions connected through junctions, runs into a scaling wall that a two-dimensional layout avoids.",{"type":21,"tag":59,"props":3427,"children":3429},{"id":3428},"why-linear-chains-are-the-default-and-why-they-cap-out",[3430],{"type":26,"value":3431},"Why linear chains are the default and why they cap out",{"type":21,"tag":22,"props":3433,"children":3434},{},[3435,3437,3442],{"type":26,"value":3436},"IonQ and Quantinuum, ",{"type":21,"tag":29,"props":3438,"children":3439},{"href":2181},[3440],{"type":26,"value":3441},"the two companies furthest along commercially",{"type":26,"value":3443}," with trapped ions, both build on one-dimensional ion chains. Ions sit in a line inside a trap, and moving ions between different zones of the chip requires junctions, physical structures where chains merge and split. Junctions work, but every one adds engineering complexity, and a chip built from linear segments scales roughly with how many segments you wire together, not with the raw area of the chip.",{"type":21,"tag":22,"props":3445,"children":3446},{},[3447],{"type":26,"value":3448},"ZuriQ's approach uses Penning micro-traps, which confine ions with static magnetic fields instead of the oscillating electric fields used in standard linear traps. According to the company, this lets ions move freely across two dimensions rather than being confined to a single line, so qubit density grows with the area of the chip rather than the number of chained segments. That is a real architectural difference, not a rebranding of the same idea, and it is the kind of claim worth watching for independent confirmation as the company scales past its current demonstrator.",{"type":21,"tag":59,"props":3450,"children":3452},{"id":3451},"what-has-been-built-so-far",[3453],{"type":26,"value":3454},"What has been built so far",{"type":21,"tag":22,"props":3456,"children":3457},{},[3458],{"type":26,"value":3459},"The concrete result behind the funding is a working 3-by-3 array of nine individually controlled ions, built in collaboration with ETH Zurich over 18 months and fabricated on Infineon's production lines. ZuriQ describes it as the largest two-dimensional trapped-ion array demonstrated to date, which is a claim about being first at a specific, small scale rather than a claim about being competitive with IonQ's or Quantinuum's qubit counts today.",{"type":21,"tag":22,"props":3461,"children":3462},{},[3463],{"type":26,"value":3464},"Nine ions is a long way from a useful quantum computer. What the demonstrator establishes is that the Penning-trap approach works in a real fabricated chip rather than only on paper, and that Infineon's existing manufacturing lines produce the hardware, which matters for whether the architecture eventually scales past a lab bench. Quantonation partner Nicolas Jurczak framed the round as evidence that real physics breakthroughs are still available in quantum hardware architecture, which is the kind of statement every early-stage investor makes about their own bet and is worth reading as exactly that.",{"type":21,"tag":59,"props":3466,"children":3468},{"id":3467},"what-this-does-and-does-not-tell-you",[3469],{"type":26,"value":3470},"What this does and does not tell you",{"type":21,"tag":22,"props":3472,"children":3473},{},[3474],{"type":26,"value":3475},"This is a seed-stage company with a nine-qubit demonstrator, not a commercial system. The claims about how the architecture scales are ZuriQ's own projections, not measured results at any meaningful qubit count. What makes it worth tracking rather than dismissing is the specificity: a named physical mechanism (Penning micro-traps), a working fabricated device, and an industrial manufacturing partner, rather than a roadmap slide with no hardware behind it.",{"type":21,"tag":22,"props":3477,"children":3478},{},[3479,3481,3486],{"type":26,"value":3480},"If ZuriQ's 2D architecture scales the way it projects, it would address one of the structural bottlenecks in ",{"type":21,"tag":29,"props":3482,"children":3483},{"href":1457},[3484],{"type":26,"value":3485},"trapped-ion scaling",{"type":26,"value":3487}," that this site has covered before: connecting qubits without every additional connection point adding its own error budget. Whether it gets there is a multi-year question. For now, the honest summary is a well-funded, technically specific bet from a team with real trapped-ion research credentials, at a scale that proves the mechanism works and nothing more.",{"title":7,"searchDepth":335,"depth":335,"links":3489},[3490,3491,3492],{"id":3428,"depth":335,"text":3431},{"id":3451,"depth":335,"text":3454},{"id":3467,"depth":335,"text":3470},"content:blog:zuriq-2d-trapped-ion-seed-funding.md","blog\u002Fzuriq-2d-trapped-ion-seed-funding.md","blog\u002Fzuriq-2d-trapped-ion-seed-funding",{"_path":3497,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":3498,"description":3499,"date":3500,"author":11,"tags":3501,"readingTime":496,"body":3503,"_type":715,"_id":4419,"_source":717,"_file":4420,"_stem":4421,"_extension":720},"\u002Fja\u002Fblog\u002Fquantum-benchmarking-eo-14413","新しい量子大統領令が明かす、この分野の本当の課題","EO 14413 を政治文書ではなく技術文書として読むと、量子コンピューティングがいまだに解決できていないことの率直なリストが見えてきます — その筆頭が、量子コンピュータの良し悪しを誰も確実に測定できないという事実です。","2026-07-18",[1033,730,3502],"Performance",{"type":18,"children":3504,"toc":4410},[3505,3517,3522,3527,3533,3538,3550,3555,3614,3619,3624,4128,4153,4159,4164,4169,4190,4202,4208,4220,4225,4251,4256,4261,4266,4271,4292,4297,4302,4307,4328,4333,4345,4350,4355,4360,4388,4393,4406],{"type":21,"tag":22,"props":3506,"children":3507},{},[3508,3510,3515],{"type":26,"value":3509},"2026年6月22日、大統領令 14413 — ",{"type":21,"tag":51,"props":3511,"children":3512},{},[3513],{"type":26,"value":3514},"Ushering in the Next Frontier of Quantum Innovation",{"type":26,"value":3516}," — が署名されました。報道はそのほぼすべてが政治的な側面に集中していました。",{"type":21,"tag":22,"props":3518,"children":3519},{},[3520],{"type":26,"value":3521},"これは残念なことです。というのも、より興味深いのは技術的な読み方だからです。この種の政策文書は、量子プログラムを実際に運営している人々からの多大な意見を取り入れて起草されます。そして彼らが要求している内容は、この分野がいまだにできないことの、かなり率直な地図になっています。そのように読めば、EO 14413 は発表というより課題リストです。",{"type":21,"tag":22,"props":3523,"children":3524},{},[3525],{"type":26,"value":3526},"以下、特に目を引く点を挙げます。",{"type":21,"tag":59,"props":3528,"children":3530},{"id":3529},"量子コンピュータの良し悪しを誰も確実に測定できない",[3531],{"type":26,"value":3532},"量子コンピュータの良し悪しを、誰も確実に測定できない",{"type":21,"tag":22,"props":3534,"children":3535},{},[3536],{"type":26,"value":3537},"この大統領令の中で最も静かに注目すべき一文は、エネルギー省に対し、180日以内に「量子コンピューティングシステムの性能を正確に評価するために必要なツールと能力を開発するための国家センター」を設立するよう指示しているものです。",{"type":21,"tag":22,"props":3539,"children":3540},{},[3541,3543,3548],{"type":26,"value":3542},"少し立ち止まって考えてみてください。Feynman が量子コンピュータを提唱してからおよそ45年、最初の量子超越性の主張から7年が経った今、政府が機関を立ち上げているのは、",{"type":21,"tag":51,"props":3544,"children":3545},{},[3546],{"type":26,"value":3547},"「この量子コンピュータは本当に優秀なのか？」",{"type":26,"value":3549}," という問いに答える信頼できる方法が我々にないからなのです。",{"type":21,"tag":22,"props":3551,"children":3552},{},[3553],{"type":26,"value":3554},"これはお役所的な前置きではありません。現実の問題であり、2つの QPU を比較しようとしたことのある人なら誰もが突き当たったことがあるはずです。問題は、ベンダーごとに報告する数値が異なることにあります。",{"type":21,"tag":3556,"props":3557,"children":3558},"ul",{},[3559,3570,3584,3594,3604],{"type":21,"tag":3560,"props":3561,"children":3562},"li",{},[3563,3568],{"type":21,"tag":71,"props":3564,"children":3565},{},[3566],{"type":26,"value":3567},"量子ビット数",{"type":26,"value":3569}," は、それ単独ではほとんど意味がありません。実用的な作業では、質の悪い100量子ビットは、質の良い20量子ビットより劣ることがあります。",{"type":21,"tag":3560,"props":3571,"children":3572},{},[3573,3582],{"type":21,"tag":71,"props":3574,"children":3575},{},[3576],{"type":21,"tag":29,"props":3577,"children":3579},{"href":3578},"\u002Fglossary\u002Fquantum-volume",[3580],{"type":26,"value":3581},"量子ボリューム",{"type":26,"value":3583}," は、量子ビット数、結合性、誤り率を1つの数値にまとめたものです — しかし飽和してしまいますし、自分が関心のある特定の回路をその装置がどう扱うかについてはほとんど何も語りません。",{"type":21,"tag":3560,"props":3585,"children":3586},{},[3587,3592],{"type":21,"tag":71,"props":3588,"children":3589},{},[3590],{"type":26,"value":3591},"ゲート忠実度",{"type":26,"value":3593}," は通常、孤立した1量子ビットゲートおよび2量子ビットゲートについて示され、これは体系的にチップを実際より良く見せてしまいます。ゲートを並列に実行すると誤りの積み重なり方が変わりますし、クロストークは見出しの数値には現れません。",{"type":21,"tag":3560,"props":3595,"children":3596},{},[3597,3602],{"type":21,"tag":71,"props":3598,"children":3599},{},[3600],{"type":26,"value":3601},"CLOPS",{"type":26,"value":3603}," などのスループット指標は速度を測るものであって、精度を測るものではありません — ノイズを高速に返すマシンは役に立ちません。",{"type":21,"tag":3560,"props":3605,"children":3606},{},[3607,3612],{"type":21,"tag":71,"props":3608,"children":3609},{},[3610],{"type":26,"value":3611},"「アルゴリズミック量子ビット」",{"type":26,"value":3613}," をはじめとするベンダー独自の指標は、ベンダー間で比較できません。そしてしばしば、それこそが狙いなのです。",{"type":21,"tag":22,"props":3615,"children":3616},{},[3617],{"type":26,"value":3618},"この大統領令はさらに、「商用量子コンピューティング能力を評価する政府の能力を向上させる」ための省庁間の情報共有の仕組みも求めています — これは、ベンダーのマーケティング上の主張を現状では独立に検証するのが難しい、と丁寧に言い換えたものです。",{"type":21,"tag":22,"props":3620,"children":3621},{},[3622],{"type":26,"value":3623},"自分でハードウェアを評価するのであれば、実践的な教訓は、単一の数値を信じるのをやめ、自分自身のワークロードでベンチマークすることです。プレスリリースに頼るのではなく、実機のデバイス特性を直接取得できます。",{"type":21,"tag":312,"props":3625,"children":3629},{"className":3626,"code":3627,"language":3628,"meta":7,"style":7},"language-python shiki shiki-themes github-dark","from qiskit_ibm_runtime import QiskitRuntimeService\n\nservice = QiskitRuntimeService()\nbackend = service.least_busy(operational=True, simulator=False)\n\nprint(f\"Backend:  {backend.name}\")\nprint(f\"Qubits:   {backend.num_qubits}\")\nprint(f\"Basis:    {backend.basis_gates}\")\n\n# Per-qubit error rates vary enormously across a single chip\nprops = backend.properties()\nerrors = [(q, props.readout_error(q)) for q in range(backend.num_qubits)]\nworst = max(errors, key=lambda x: x[1])\nbest = min(errors, key=lambda x: x[1])\nprint(f\"Readout error — best qubit: {best[1]:.4f}, worst: {worst[1]:.4f}\")\n","python",[3630],{"type":21,"tag":79,"props":3631,"children":3632},{"__ignoreMap":7},[3633,3657,3666,3684,3739,3746,3793,3834,3875,3882,3892,3910,3952,4000,4042],{"type":21,"tag":322,"props":3634,"children":3635},{"class":324,"line":325},[3636,3642,3647,3652],{"type":21,"tag":322,"props":3637,"children":3639},{"style":3638},"--shiki-default:#F97583",[3640],{"type":26,"value":3641},"from",{"type":21,"tag":322,"props":3643,"children":3644},{"style":329},[3645],{"type":26,"value":3646}," qiskit_ibm_runtime ",{"type":21,"tag":322,"props":3648,"children":3649},{"style":3638},[3650],{"type":26,"value":3651},"import",{"type":21,"tag":322,"props":3653,"children":3654},{"style":329},[3655],{"type":26,"value":3656}," QiskitRuntimeService\n",{"type":21,"tag":322,"props":3658,"children":3659},{"class":324,"line":335},[3660],{"type":21,"tag":322,"props":3661,"children":3663},{"emptyLinePlaceholder":3662},true,[3664],{"type":26,"value":3665},"\n",{"type":21,"tag":322,"props":3667,"children":3668},{"class":324,"line":344},[3669,3674,3679],{"type":21,"tag":322,"props":3670,"children":3671},{"style":329},[3672],{"type":26,"value":3673},"service ",{"type":21,"tag":322,"props":3675,"children":3676},{"style":3638},[3677],{"type":26,"value":3678},"=",{"type":21,"tag":322,"props":3680,"children":3681},{"style":329},[3682],{"type":26,"value":3683}," QiskitRuntimeService()\n",{"type":21,"tag":322,"props":3685,"children":3686},{"class":324,"line":407},[3687,3692,3696,3701,3707,3711,3716,3720,3725,3729,3734],{"type":21,"tag":322,"props":3688,"children":3689},{"style":329},[3690],{"type":26,"value":3691},"backend ",{"type":21,"tag":322,"props":3693,"children":3694},{"style":3638},[3695],{"type":26,"value":3678},{"type":21,"tag":322,"props":3697,"children":3698},{"style":329},[3699],{"type":26,"value":3700}," service.least_busy(",{"type":21,"tag":322,"props":3702,"children":3704},{"style":3703},"--shiki-default:#FFAB70",[3705],{"type":26,"value":3706},"operational",{"type":21,"tag":322,"props":3708,"children":3709},{"style":3638},[3710],{"type":26,"value":3678},{"type":21,"tag":322,"props":3712,"children":3713},{"style":348},[3714],{"type":26,"value":3715},"True",{"type":21,"tag":322,"props":3717,"children":3718},{"style":329},[3719],{"type":26,"value":367},{"type":21,"tag":322,"props":3721,"children":3722},{"style":3703},[3723],{"type":26,"value":3724},"simulator",{"type":21,"tag":322,"props":3726,"children":3727},{"style":3638},[3728],{"type":26,"value":3678},{"type":21,"tag":322,"props":3730,"children":3731},{"style":348},[3732],{"type":26,"value":3733},"False",{"type":21,"tag":322,"props":3735,"children":3736},{"style":329},[3737],{"type":26,"value":3738},")\n",{"type":21,"tag":322,"props":3740,"children":3741},{"class":324,"line":431},[3742],{"type":21,"tag":322,"props":3743,"children":3744},{"emptyLinePlaceholder":3662},[3745],{"type":26,"value":3665},{"type":21,"tag":322,"props":3747,"children":3748},{"class":324,"line":453},[3749,3754,3759,3764,3769,3774,3779,3784,3789],{"type":21,"tag":322,"props":3750,"children":3751},{"style":348},[3752],{"type":26,"value":3753},"print",{"type":21,"tag":322,"props":3755,"children":3756},{"style":329},[3757],{"type":26,"value":3758},"(",{"type":21,"tag":322,"props":3760,"children":3761},{"style":3638},[3762],{"type":26,"value":3763},"f",{"type":21,"tag":322,"props":3765,"children":3766},{"style":359},[3767],{"type":26,"value":3768},"\"Backend:  ",{"type":21,"tag":322,"props":3770,"children":3771},{"style":348},[3772],{"type":26,"value":3773},"{",{"type":21,"tag":322,"props":3775,"children":3776},{"style":329},[3777],{"type":26,"value":3778},"backend.name",{"type":21,"tag":322,"props":3780,"children":3781},{"style":348},[3782],{"type":26,"value":3783},"}",{"type":21,"tag":322,"props":3785,"children":3786},{"style":359},[3787],{"type":26,"value":3788},"\"",{"type":21,"tag":322,"props":3790,"children":3791},{"style":329},[3792],{"type":26,"value":3738},{"type":21,"tag":322,"props":3794,"children":3795},{"class":324,"line":474},[3796,3800,3804,3808,3813,3817,3822,3826,3830],{"type":21,"tag":322,"props":3797,"children":3798},{"style":348},[3799],{"type":26,"value":3753},{"type":21,"tag":322,"props":3801,"children":3802},{"style":329},[3803],{"type":26,"value":3758},{"type":21,"tag":322,"props":3805,"children":3806},{"style":3638},[3807],{"type":26,"value":3763},{"type":21,"tag":322,"props":3809,"children":3810},{"style":359},[3811],{"type":26,"value":3812},"\"Qubits:   ",{"type":21,"tag":322,"props":3814,"children":3815},{"style":348},[3816],{"type":26,"value":3773},{"type":21,"tag":322,"props":3818,"children":3819},{"style":329},[3820],{"type":26,"value":3821},"backend.num_qubits",{"type":21,"tag":322,"props":3823,"children":3824},{"style":348},[3825],{"type":26,"value":3783},{"type":21,"tag":322,"props":3827,"children":3828},{"style":359},[3829],{"type":26,"value":3788},{"type":21,"tag":322,"props":3831,"children":3832},{"style":329},[3833],{"type":26,"value":3738},{"type":21,"tag":322,"props":3835,"children":3836},{"class":324,"line":496},[3837,3841,3845,3849,3854,3858,3863,3867,3871],{"type":21,"tag":322,"props":3838,"children":3839},{"style":348},[3840],{"type":26,"value":3753},{"type":21,"tag":322,"props":3842,"children":3843},{"style":329},[3844],{"type":26,"value":3758},{"type":21,"tag":322,"props":3846,"children":3847},{"style":3638},[3848],{"type":26,"value":3763},{"type":21,"tag":322,"props":3850,"children":3851},{"style":359},[3852],{"type":26,"value":3853},"\"Basis:    ",{"type":21,"tag":322,"props":3855,"children":3856},{"style":348},[3857],{"type":26,"value":3773},{"type":21,"tag":322,"props":3859,"children":3860},{"style":329},[3861],{"type":26,"value":3862},"backend.basis_gates",{"type":21,"tag":322,"props":3864,"children":3865},{"style":348},[3866],{"type":26,"value":3783},{"type":21,"tag":322,"props":3868,"children":3869},{"style":359},[3870],{"type":26,"value":3788},{"type":21,"tag":322,"props":3872,"children":3873},{"style":329},[3874],{"type":26,"value":3738},{"type":21,"tag":322,"props":3876,"children":3877},{"class":324,"line":16},[3878],{"type":21,"tag":322,"props":3879,"children":3880},{"emptyLinePlaceholder":3662},[3881],{"type":26,"value":3665},{"type":21,"tag":322,"props":3883,"children":3885},{"class":324,"line":3884},10,[3886],{"type":21,"tag":322,"props":3887,"children":3889},{"style":3888},"--shiki-default:#6A737D",[3890],{"type":26,"value":3891},"# Per-qubit error rates vary enormously across a single chip\n",{"type":21,"tag":322,"props":3893,"children":3895},{"class":324,"line":3894},11,[3896,3901,3905],{"type":21,"tag":322,"props":3897,"children":3898},{"style":329},[3899],{"type":26,"value":3900},"props ",{"type":21,"tag":322,"props":3902,"children":3903},{"style":3638},[3904],{"type":26,"value":3678},{"type":21,"tag":322,"props":3906,"children":3907},{"style":329},[3908],{"type":26,"value":3909}," backend.properties()\n",{"type":21,"tag":322,"props":3911,"children":3912},{"class":324,"line":1034},[3913,3918,3922,3927,3932,3937,3942,3947],{"type":21,"tag":322,"props":3914,"children":3915},{"style":329},[3916],{"type":26,"value":3917},"errors ",{"type":21,"tag":322,"props":3919,"children":3920},{"style":3638},[3921],{"type":26,"value":3678},{"type":21,"tag":322,"props":3923,"children":3924},{"style":329},[3925],{"type":26,"value":3926}," [(q, props.readout_error(q)) ",{"type":21,"tag":322,"props":3928,"children":3929},{"style":3638},[3930],{"type":26,"value":3931},"for",{"type":21,"tag":322,"props":3933,"children":3934},{"style":329},[3935],{"type":26,"value":3936}," q ",{"type":21,"tag":322,"props":3938,"children":3939},{"style":3638},[3940],{"type":26,"value":3941},"in",{"type":21,"tag":322,"props":3943,"children":3944},{"style":348},[3945],{"type":26,"value":3946}," range",{"type":21,"tag":322,"props":3948,"children":3949},{"style":329},[3950],{"type":26,"value":3951},"(backend.num_qubits)]\n",{"type":21,"tag":322,"props":3953,"children":3955},{"class":324,"line":3954},13,[3956,3961,3965,3970,3975,3980,3985,3990,3995],{"type":21,"tag":322,"props":3957,"children":3958},{"style":329},[3959],{"type":26,"value":3960},"worst ",{"type":21,"tag":322,"props":3962,"children":3963},{"style":3638},[3964],{"type":26,"value":3678},{"type":21,"tag":322,"props":3966,"children":3967},{"style":348},[3968],{"type":26,"value":3969}," max",{"type":21,"tag":322,"props":3971,"children":3972},{"style":329},[3973],{"type":26,"value":3974},"(errors, ",{"type":21,"tag":322,"props":3976,"children":3977},{"style":3703},[3978],{"type":26,"value":3979},"key",{"type":21,"tag":322,"props":3981,"children":3982},{"style":3638},[3983],{"type":26,"value":3984},"=lambda",{"type":21,"tag":322,"props":3986,"children":3987},{"style":329},[3988],{"type":26,"value":3989}," x: x[",{"type":21,"tag":322,"props":3991,"children":3992},{"style":348},[3993],{"type":26,"value":3994},"1",{"type":21,"tag":322,"props":3996,"children":3997},{"style":329},[3998],{"type":26,"value":3999},"])\n",{"type":21,"tag":322,"props":4001,"children":4003},{"class":324,"line":4002},14,[4004,4009,4013,4018,4022,4026,4030,4034,4038],{"type":21,"tag":322,"props":4005,"children":4006},{"style":329},[4007],{"type":26,"value":4008},"best ",{"type":21,"tag":322,"props":4010,"children":4011},{"style":3638},[4012],{"type":26,"value":3678},{"type":21,"tag":322,"props":4014,"children":4015},{"style":348},[4016],{"type":26,"value":4017}," min",{"type":21,"tag":322,"props":4019,"children":4020},{"style":329},[4021],{"type":26,"value":3974},{"type":21,"tag":322,"props":4023,"children":4024},{"style":3703},[4025],{"type":26,"value":3979},{"type":21,"tag":322,"props":4027,"children":4028},{"style":3638},[4029],{"type":26,"value":3984},{"type":21,"tag":322,"props":4031,"children":4032},{"style":329},[4033],{"type":26,"value":3989},{"type":21,"tag":322,"props":4035,"children":4036},{"style":348},[4037],{"type":26,"value":3994},{"type":21,"tag":322,"props":4039,"children":4040},{"style":329},[4041],{"type":26,"value":3999},{"type":21,"tag":322,"props":4043,"children":4045},{"class":324,"line":4044},15,[4046,4050,4054,4058,4063,4067,4072,4076,4081,4086,4090,4095,4099,4104,4108,4112,4116,4120,4124],{"type":21,"tag":322,"props":4047,"children":4048},{"style":348},[4049],{"type":26,"value":3753},{"type":21,"tag":322,"props":4051,"children":4052},{"style":329},[4053],{"type":26,"value":3758},{"type":21,"tag":322,"props":4055,"children":4056},{"style":3638},[4057],{"type":26,"value":3763},{"type":21,"tag":322,"props":4059,"children":4060},{"style":359},[4061],{"type":26,"value":4062},"\"Readout error — best qubit: ",{"type":21,"tag":322,"props":4064,"children":4065},{"style":348},[4066],{"type":26,"value":3773},{"type":21,"tag":322,"props":4068,"children":4069},{"style":329},[4070],{"type":26,"value":4071},"best[",{"type":21,"tag":322,"props":4073,"children":4074},{"style":348},[4075],{"type":26,"value":3994},{"type":21,"tag":322,"props":4077,"children":4078},{"style":329},[4079],{"type":26,"value":4080},"]",{"type":21,"tag":322,"props":4082,"children":4083},{"style":3638},[4084],{"type":26,"value":4085},":.4f",{"type":21,"tag":322,"props":4087,"children":4088},{"style":348},[4089],{"type":26,"value":3783},{"type":21,"tag":322,"props":4091,"children":4092},{"style":359},[4093],{"type":26,"value":4094},", worst: ",{"type":21,"tag":322,"props":4096,"children":4097},{"style":348},[4098],{"type":26,"value":3773},{"type":21,"tag":322,"props":4100,"children":4101},{"style":329},[4102],{"type":26,"value":4103},"worst[",{"type":21,"tag":322,"props":4105,"children":4106},{"style":348},[4107],{"type":26,"value":3994},{"type":21,"tag":322,"props":4109,"children":4110},{"style":329},[4111],{"type":26,"value":4080},{"type":21,"tag":322,"props":4113,"children":4114},{"style":3638},[4115],{"type":26,"value":4085},{"type":21,"tag":322,"props":4117,"children":4118},{"style":348},[4119],{"type":26,"value":3783},{"type":21,"tag":322,"props":4121,"children":4122},{"style":359},[4123],{"type":26,"value":3788},{"type":21,"tag":322,"props":4125,"children":4126},{"style":329},[4127],{"type":26,"value":3738},{"type":21,"tag":22,"props":4129,"children":4130},{},[4131,4133,4138,4140,4145,4146,4151],{"type":26,"value":4132},"これを現行のほぼどのデバイスで実行しても、",{"type":21,"tag":51,"props":4134,"children":4135},{},[4136],{"type":26,"value":4137},"同じチップ上",{"type":26,"value":4139}," での最良の量子ビットと最悪の量子ビットの差が、5倍以上になることが多いと分かるでしょう。この差は、あらゆる要約統計量からは見えません。だからこそ、トランスパイルと量子ビットの選択がこれほど重要になるのです。バックエンドの選び方については、",{"type":21,"tag":29,"props":4141,"children":4142},{"href":814},[4143],{"type":26,"value":4144},"SDK 比較",{"type":26,"value":811},{"type":21,"tag":29,"props":4147,"children":4148},{"href":806},[4149],{"type":26,"value":4150},"ハードウェアガイド",{"type":26,"value":4152},"でさらに詳しく解説しています。",{"type":21,"tag":59,"props":4154,"children":4156},{"id":4155},"この分野は勝者となる量子ビット技術をまだ選んでいない",[4157],{"type":26,"value":4158},"この分野は、勝者となる量子ビット技術をまだ選んでいない",{"type":21,"tag":22,"props":4160,"children":4161},{},[4162],{"type":26,"value":4163},"その不在が注目に値するものがあります。この大統領令は、量子ビットの方式を一度も名指ししていません。超伝導回路、イオントラップ、光量子、中性原子、スピン量子ビットのいずれにも優先順位を付けていません。総称的に「量子を実現する構成要素技術」と述べるにとどまっています。",{"type":21,"tag":22,"props":4165,"children":4166},{},[4167],{"type":26,"value":4168},"これは意味のあるシグナルです。技術が成熟すると、調達文書は具体的になります。「何らかのスイッチング素子に基づく計算装置」のための国家戦略を書く人はいません。この意図的な曖昧さは、紛れもない現実を反映しています。2026年半ば時点で、明確に勝利した方式は存在しないのです。",{"type":21,"tag":22,"props":4170,"children":4171},{},[4172,4174,4180,4182,4188],{"type":26,"value":4173},"それぞれが異なる軸で先行しています — 超伝導チップはゲート速度と製造スケール、イオントラップは忠実度と全対全結合、中性原子は量子ビット数、光量子は室温動作とネットワーク接続性です。2026年の",{"type":21,"tag":29,"props":4175,"children":4177},{"href":4176},"\u002Fresearch\u002Fholographic-codes-quantum-gravity-2026",[4178],{"type":26,"value":4179},"ホログラフィック符号の実験",{"type":26,"value":4181},"はイオントラップ上で行われ、",{"type":21,"tag":29,"props":4183,"children":4185},{"href":4184},"\u002Fresearch\u002Fgoogle-below-threshold-2024",[4186],{"type":26,"value":4187},"Google の閾値以下の誤り訂正の成果",{"type":26,"value":4189},"は超伝導ハードウェア上で行われました。どちらも画期的な成果ですが、まったく異なるプラットフォーム上のものでした。",{"type":21,"tag":22,"props":4191,"children":4192},{},[4193,4195,4200],{"type":26,"value":4194},"量子コンピューティングを学ぶ人にとって、これは実は朗報であり、具体的な含意があります。",{"type":21,"tag":71,"props":4196,"children":4197},{},[4198],{"type":26,"value":4199},"特定ベンダーのスタックに過度に特化しないこと。",{"type":26,"value":4201}," 回路、ゲート、測定、誤り緩和といった抽象概念はハードウェアを越えて通用します。ベンダー固有の API の詳細は、そうとは限りません。",{"type":21,"tag":59,"props":4203,"children":4205},{"id":4204},"いまや基準は現行の古典的能力を超えること",[4206],{"type":26,"value":4207},"いまや基準は「現行の古典的能力を超えること」",{"type":21,"tag":22,"props":4209,"children":4210},{},[4211,4213,4218],{"type":26,"value":4212},"この大統領令は、",{"type":21,"tag":71,"props":4214,"children":4215},{},[4216],{"type":26,"value":4217},"Quantum Computer for Application Development and Discovery Science (QC-ADDS)",{"type":26,"value":4219}," の取り組みを立ち上げ、少なくとも1台のマシンをエネルギー省の施設に納入し、科学コミュニティが利用できるようにすることを目指しています。エネルギー長官は90日以内に、「変革的な科学応用……経済的に重要な応用へ、そして現行の古典コンピュータの能力を超える道筋の上にある」システムの仕様を特定しなければなりません。",{"type":21,"tag":22,"props":4221,"children":4222},{},[4223],{"type":26,"value":4224},"この言い回しには、注目すべき点が2つあります。",{"type":21,"tag":22,"props":4226,"children":4227},{},[4228,4230,4235,4237,4242,4244,4249],{"type":26,"value":4229},"第一に、その枠組みは商業的な優位性ではなく ",{"type":21,"tag":71,"props":4231,"children":4232},{},[4233],{"type":26,"value":4234},"ディスカバリーサイエンス（発見のための科学）",{"type":26,"value":4236}," です。目標は、ビジネス上のタスクで古典コンピュータを打ち負かす製品ではなく、研究者が物事を知るために使う科学的な計測器です。これは実際に証拠が指し示す方向と一致しています。",{"type":21,"tag":29,"props":4238,"children":4239},{"href":2054},[4240],{"type":26,"value":4241},"IBM の2023年の量子ユーティリティの成果",{"type":26,"value":4243},"も、ホログラフィック符号の研究も、どちらも",{"type":21,"tag":51,"props":4245,"children":4246},{},[4247],{"type":26,"value":4248},"物理学",{"type":26,"value":4250},"の実験でした。",{"type":21,"tag":22,"props":4252,"children":4253},{},[4254],{"type":26,"value":4255},"第二に、「〜への道筋の上にある」という表現は、この文中で実質的な役割を果たしています。仕様が定められようとしているマシンは到達点ではなく一段階だ、という認識の表れです。",{"type":21,"tag":59,"props":4257,"children":4259},{"id":4258},"分散量子コンピューティングが真剣な目標として登場する",[4260],{"type":26,"value":4258},{"type":21,"tag":22,"props":4262,"children":4263},{},[4264],{"type":26,"value":4265},"ネットワーキングに関する節に紛れ込んでいるのが、「分散量子コンピューティング」のための量子ネットワークを対象とした DOE の計画への要請です。",{"type":21,"tag":22,"props":4267,"children":4268},{},[4269],{"type":26,"value":4270},"これが重要なのは、スケーリングについての告白だからです。1台の巨大な量子プロセッサを構築するのは、とてつもなく困難です — 配線、冷却、クロストーク、歩留まりのすべてが、チップが大きくなるほど悪化します。複数の小さなプロセッサを1つの論理的なマシンに連結するのは代替となる道であり、それは別々の装置の間で、実用に足るだけの高い忠実度でもつれを分配できるかにかかっています。",{"type":21,"tag":22,"props":4272,"children":4273},{},[4274,4276,4282,4284,4290],{"type":26,"value":4275},"その基礎となるプリミティブが",{"type":21,"tag":29,"props":4277,"children":4279},{"href":4278},"\u002Fresearch\u002Fbennett-teleportation-1993",[4280],{"type":26,"value":4281},"量子テレポーテーション",{"type":26,"value":4283},"で、共有された",{"type":21,"tag":29,"props":4285,"children":4287},{"href":4286},"\u002Fglossary\u002Fentanglement",[4288],{"type":26,"value":4289},"もつれ",{"type":26,"value":4291},"と2ビットの古典情報を使って、量子ビットの状態を場所から場所へ移動させます。実験的には何年も前から日常的に行われていますが — プロセッサ同士を編み上げられるほど高速に、かつ十分にクリーンに行うのは、そうではありません。これが5年間の計画目標として名指しされていることは、単一の巨大チップという戦略だけでは十分だと想定されていない証拠です。",{"type":21,"tag":59,"props":4293,"children":4295},{"id":4294},"人材ギャップがボトルネックとして扱われている",[4296],{"type":26,"value":4294},{"type":21,"tag":22,"props":4298,"children":4299},{},[4300],{"type":26,"value":4301},"この大統領令は NSF に対し、180日以内に「全米 QIST 人材育成研究所のネットワーク」を立ち上げるよう、採用・定着戦略とあわせて指示しています。",{"type":21,"tag":22,"props":4303,"children":4304},{},[4305],{"type":26,"value":4306},"人材に関する節は読み飛ばしがちですが、その記載は具体的なことを教えてくれます。量子分野の進展を縛る制約は、ハードウェアだけではないということです。量子プログラムを書き、デバッグし、その挙動を筋道立てて考えられる人材が足りていません — そしてその不足は今や、専用の機関を設けるに値するほど深刻だと見なされているのです。",{"type":21,"tag":22,"props":4308,"children":4309},{},[4310,4312,4318,4320,4326],{"type":26,"value":4311},"このサイトを読んでいる方にとって、これは文書全体の中で最も直接的に行動につながる部分です。希少だと言われているスキルは、今すぐ、無料で、実機のハードウェア上で習得できます。",{"type":21,"tag":29,"props":4313,"children":4315},{"href":4314},"\u002Fcourses",[4316],{"type":26,"value":4317},"コースページ",{"type":26,"value":4319},"には最良の体系的な選択肢をまとめてありますし、",{"type":21,"tag":29,"props":4321,"children":4323},{"href":4322},"\u002Fblog\u002Fgetting-started-free-quantum-computing",[4324],{"type":26,"value":4325},"入門ガイド",{"type":26,"value":4327},"を読めば、午後のひとときで実機の QPU 上に回路を走らせられるようになります。",{"type":21,"tag":59,"props":4329,"children":4331},{"id":4330},"この大統領令に含まれていないもの",[4332],{"type":26,"value":4330},{"type":21,"tag":22,"props":4334,"children":4335},{},[4336,4338,4343],{"type":26,"value":4337},"正確を期すために、はっきり述べておく価値があります。この大統領令には ",{"type":21,"tag":71,"props":4339,"children":4340},{},[4341],{"type":26,"value":4342},"金額の記載が一切ありません",{"type":26,"value":4344},"。方向性を定め、責任を割り当て、期限を課してはいます — 各節にわたって90日、120日、180日、210日といった具合に — が、予算の充当は別のところから来ます。また、大統領令の期限は成果が出ることの保証でもありません。",{"type":21,"tag":22,"props":4346,"children":4347},{},[4348],{"type":26,"value":4349},"タイムラインは意図の表明として捉え、それを前提に計画を立てられるスケジュールとは考えないでください。",{"type":21,"tag":59,"props":4351,"children":4353},{"id":4352},"まとめ",[4354],{"type":26,"value":4352},{"type":21,"tag":22,"props":4356,"children":4357},{},[4358],{"type":26,"value":4359},"政治的な要素を取り除けば、EO 14413 は異例なほど率直な技術評価として読めます。",{"type":21,"tag":3556,"props":4361,"children":4362},{},[4363,4368,4373,4378,4383],{"type":21,"tag":3560,"props":4364,"children":4365},{},[4366],{"type":26,"value":4367},"我々は量子コンピュータの性能を確実に測定できず、それは今や機関化を要するほど緊急の課題である。",{"type":21,"tag":3560,"props":4369,"children":4370},{},[4371],{"type":26,"value":4372},"勝利した量子ビット技術はまだなく、賭けは分散されている。",{"type":21,"tag":3560,"props":4374,"children":4375},{},[4376],{"type":26,"value":4377},"短期的な目標は商業的優位性ではなく、科学的発見である。",{"type":21,"tag":3560,"props":4379,"children":4380},{},[4381],{"type":26,"value":4382},"スケーリングには、1台の巨大なチップを作るのではなく、チップ同士をネットワーク接続することが必要かもしれない。",{"type":21,"tag":3560,"props":4384,"children":4385},{},[4386],{"type":26,"value":4387},"訓練された人材が足りておらず、それは一次的なボトルネックである。",{"type":21,"tag":22,"props":4389,"children":4390},{},[4391],{"type":26,"value":4392},"これらのどれも悲観的ではありません。「そもそもこれは動くのか」を通り越して、「どう測り、どうスケールさせ、どう人を配置するか」へ進んだ分野の姿です — これは、古典コンピューティングが1950年代に経た移行とおおむね同じものです。",{"type":21,"tag":22,"props":4394,"children":4395},{},[4396,4398,4404],{"type":26,"value":4397},"注目すべきは測定の問題です。ベンチマークは量子ビット数の記録に比べれば地味に聞こえますが、測定できないものは設計できませんし、量子優位性についての真剣な主張はすべて、最終的にはそこに立脚します。この分野が実際にどこへ向かっているのかを理解したいなら、見出しではなく誤り率を読めるようになりましょう — まずは",{"type":21,"tag":29,"props":4399,"children":4401},{"href":4400},"\u002Fglossary",[4402],{"type":26,"value":4403},"重要な用語の用語集",{"type":26,"value":4405},"から始めてください。",{"type":21,"tag":703,"props":4407,"children":4408},{},[4409],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":4411},[4412,4413,4414,4415,4416,4417,4418],{"id":3529,"depth":335,"text":3532},{"id":4155,"depth":335,"text":4158},{"id":4204,"depth":335,"text":4207},{"id":4258,"depth":335,"text":4258},{"id":4294,"depth":335,"text":4294},{"id":4330,"depth":335,"text":4330},{"id":4352,"depth":335,"text":4352},"content:ja:blog:quantum-benchmarking-eo-14413.md","ja\u002Fblog\u002Fquantum-benchmarking-eo-14413.md","ja\u002Fblog\u002Fquantum-benchmarking-eo-14413",{"_path":4423,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":4424,"description":4425,"date":3500,"author":11,"tags":4426,"readingTime":3884,"body":4427,"_type":715,"_id":7577,"_source":717,"_file":7578,"_stem":7579,"_extension":720},"\u002Fblog\u002Ferror-mitigation-practical-guide","Quantum Error Mitigation: A Practical Guide (And When It Lies to You)","Hands-on error mitigation for NISQ hardware: readout calibration, zero-noise extrapolation, Pauli twirling, and dynamical decoupling, plus the ZNE failure mode that manufactures fake improvements.",[729,1962,3502],{"type":18,"children":4428,"toc":7566},[4429,4434,4439,4445,4450,4476,4497,4502,4617,4636,4642,4647,4692,6085,6095,6103,6131,6158,6164,6185,6198,6393,6398,6404,6416,6421,6433,6439,6451,6479,7221,7231,7239,7244,7250,7255,7263,7275,7302,7314,7324,7330,7335,7353,7385,7401,7407,7419,7432,7438,7443,7497,7521,7534,7562],{"type":21,"tag":22,"props":4430,"children":4431},{},[4432],{"type":26,"value":4433},"You ran your circuit on real hardware. The answer is wrong. Not catastrophically wrong. Wrong enough to be useless. Error mitigation is the set of tricks that gets you from \"wrong\" to \"usable\" without waiting a decade for fault tolerance.",{"type":21,"tag":22,"props":4435,"children":4436},{},[4437],{"type":26,"value":4438},"This guide covers what works, in the order you should try it, with runnable Qiskit code. It also covers the part most tutorials skip: how the most popular mitigation technique sometimes produces a convincing improvement that is entirely fictitious, and the one-line sanity check that catches it.",{"type":21,"tag":59,"props":4440,"children":4442},{"id":4441},"mitigation-is-not-correction",[4443],{"type":26,"value":4444},"Mitigation Is Not Correction",{"type":21,"tag":22,"props":4446,"children":4447},{},[4448],{"type":26,"value":4449},"These get conflated constantly, so let's be precise.",{"type":21,"tag":22,"props":4451,"children":4452},{},[4453,4460,4462,4467,4469,4474],{"type":21,"tag":71,"props":4454,"children":4455},{},[4456],{"type":21,"tag":29,"props":4457,"children":4458},{"href":1526},[4459],{"type":26,"value":1529},{"type":26,"value":4461}," detects and fixes errors ",{"type":21,"tag":51,"props":4463,"children":4464},{},[4465],{"type":26,"value":4466},"during",{"type":26,"value":4468}," the computation. It encodes one logical qubit across many physical qubits, measures syndromes with ancillas, and applies corrections in real time. It preserves arbitrary quantum states to arbitrary depth, but it demands hardware below the fault-tolerance threshold and roughly 1,000 times qubit overhead. Our ",{"type":21,"tag":29,"props":4470,"children":4471},{"href":854},[4472],{"type":26,"value":4473},"error correction primer",{"type":26,"value":4475}," walks through how the surface code does this.",{"type":21,"tag":22,"props":4477,"children":4478},{},[4479,4488,4490,4495],{"type":21,"tag":71,"props":4480,"children":4481},{},[4482],{"type":21,"tag":29,"props":4483,"children":4485},{"href":4484},"\u002Fglossary\u002Ferror-mitigation",[4486],{"type":26,"value":4487},"Error mitigation",{"type":26,"value":4489}," does nothing during the computation. It runs the noisy circuit, collects noisy results, and applies classical statistics afterwards to ",{"type":21,"tag":51,"props":4491,"children":4492},{},[4493],{"type":26,"value":4494},"estimate",{"type":26,"value":4496}," what a noiseless machine would have said.",{"type":21,"tag":22,"props":4498,"children":4499},{},[4500],{"type":26,"value":4501},"The practical differences matter:",{"type":21,"tag":4503,"props":4504,"children":4505},"table",{},[4506,4528],{"type":21,"tag":4507,"props":4508,"children":4509},"thead",{},[4510],{"type":21,"tag":4511,"props":4512,"children":4513},"tr",{},[4514,4518,4523],{"type":21,"tag":4515,"props":4516,"children":4517},"th",{},[],{"type":21,"tag":4515,"props":4519,"children":4520},{},[4521],{"type":26,"value":4522},"Correction",{"type":21,"tag":4515,"props":4524,"children":4525},{},[4526],{"type":26,"value":4527},"Mitigation",{"type":21,"tag":4529,"props":4530,"children":4531},"tbody",{},[4532,4563,4581,4599],{"type":21,"tag":4511,"props":4533,"children":4534},{},[4535,4541,4552],{"type":21,"tag":4536,"props":4537,"children":4538},"td",{},[4539],{"type":26,"value":4540},"Cost",{"type":21,"tag":4536,"props":4542,"children":4543},{},[4544,4546,4550],{"type":26,"value":4545},"Extra ",{"type":21,"tag":71,"props":4547,"children":4548},{},[4549],{"type":26,"value":1797},{"type":26,"value":4551}," (~1000×)",{"type":21,"tag":4536,"props":4553,"children":4554},{},[4555,4556,4561],{"type":26,"value":4545},{"type":21,"tag":71,"props":4557,"children":4558},{},[4559],{"type":26,"value":4560},"shots",{"type":26,"value":4562}," (2–100×)",{"type":21,"tag":4511,"props":4564,"children":4565},{},[4566,4571,4576],{"type":21,"tag":4536,"props":4567,"children":4568},{},[4569],{"type":26,"value":4570},"Recovers",{"type":21,"tag":4536,"props":4572,"children":4573},{},[4574],{"type":26,"value":4575},"The quantum state",{"type":21,"tag":4536,"props":4577,"children":4578},{},[4579],{"type":26,"value":4580},"An expectation value",{"type":21,"tag":4511,"props":4582,"children":4583},{},[4584,4589,4594],{"type":21,"tag":4536,"props":4585,"children":4586},{},[4587],{"type":26,"value":4588},"Scales to",{"type":21,"tag":4536,"props":4590,"children":4591},{},[4592],{"type":26,"value":4593},"Arbitrary depth",{"type":21,"tag":4536,"props":4595,"children":4596},{},[4597],{"type":26,"value":4598},"Shallow-to-moderate depth only",{"type":21,"tag":4511,"props":4600,"children":4601},{},[4602,4607,4612],{"type":21,"tag":4536,"props":4603,"children":4604},{},[4605],{"type":26,"value":4606},"Available today",{"type":21,"tag":4536,"props":4608,"children":4609},{},[4610],{"type":26,"value":4611},"Barely",{"type":21,"tag":4536,"props":4613,"children":4614},{},[4615],{"type":26,"value":4616},"Yes, right now",{"type":21,"tag":22,"props":4618,"children":4619},{},[4620,4622,4627,4629,4634],{"type":26,"value":4621},"That last row is why mitigation dominates ",{"type":21,"tag":29,"props":4623,"children":4624},{"href":263},[4625],{"type":26,"value":4626},"NISQ-era",{"type":26,"value":4628}," work. But note row three: mitigation's overhead grows ",{"type":21,"tag":51,"props":4630,"children":4631},{},[4632],{"type":26,"value":4633},"exponentially",{"type":26,"value":4635}," with circuit depth and error rate. There is a depth beyond which no amount of shots rescues you. Knowing where that wall is (for your circuit, on your device) is the entire skill.",{"type":21,"tag":59,"props":4637,"children":4639},{"id":4638},"step-1-readout-error-mitigation-do-this-first",[4640],{"type":26,"value":4641},"Step 1: Readout Error Mitigation (Do This First)",{"type":21,"tag":22,"props":4643,"children":4644},{},[4645],{"type":26,"value":4646},"Measurement errors are the cheapest thing to fix and often the biggest single contributor. A superconducting qubit typically misreads 1–5% of the time. On 4 qubits at 3% each, you're losing ~11% of your signal before any gate error is counted.",{"type":21,"tag":22,"props":4648,"children":4649},{},[4650,4652,4658,4660,4666,4668,4674,4676,4682,4684,4690],{"type":26,"value":4651},"The fix: characterize the readout with calibration circuits, then invert. Prepare each computational basis state, measure it, and record where the counts land. That gives you an assignment matrix ",{"type":21,"tag":79,"props":4653,"children":4655},{"className":4654},[],[4656],{"type":26,"value":4657},"M",{"type":26,"value":4659}," where ",{"type":21,"tag":79,"props":4661,"children":4663},{"className":4662},[],[4664],{"type":26,"value":4665},"M[i][j]",{"type":26,"value":4667}," is the probability of reading ",{"type":21,"tag":79,"props":4669,"children":4671},{"className":4670},[],[4672],{"type":26,"value":4673},"i",{"type":26,"value":4675}," when the true state was ",{"type":21,"tag":79,"props":4677,"children":4679},{"className":4678},[],[4680],{"type":26,"value":4681},"j",{"type":26,"value":4683},". Your observed distribution is ",{"type":21,"tag":79,"props":4685,"children":4687},{"className":4686},[],[4688],{"type":26,"value":4689},"M @ p_true",{"type":26,"value":4691},", so apply the pseudo-inverse.",{"type":21,"tag":312,"props":4693,"children":4695},{"code":4694,"language":3628,"meta":7,"className":3626,"style":7},"import numpy as np\nfrom qiskit import QuantumCircuit, transpile\nfrom qiskit_aer import AerSimulator\nfrom qiskit_aer.noise import NoiseModel, depolarizing_error, ReadoutError\n\ndef build_noise(p1=0.002, p2=0.02, p_read=0.05):\n    nm = NoiseModel()\n    nm.add_all_qubit_quantum_error(depolarizing_error(p1, 1), [\"rz\", \"sx\", \"x\", \"h\"])\n    nm.add_all_qubit_quantum_error(depolarizing_error(p2, 2), [\"cx\", \"cz\", \"ecr\"])\n    nm.add_all_qubit_readout_error(\n        ReadoutError([[1 - p_read, p_read], [p_read, 1 - p_read]])\n    )\n    return nm\n\nbackend = AerSimulator(noise_model=build_noise())\nSHOTS = 40000\n\ndef calibration_matrix(n, backend, shots=20000):\n    M = np.zeros((2**n, 2**n))\n    for i in range(2**n):\n        qc = QuantumCircuit(n, n)\n        bits = format(i, f\"0{n}b\")[::-1]      # Qiskit is little-endian\n        for q, b in enumerate(bits):\n            if b == \"1\":\n                qc.x(q)\n        qc.measure(range(n), range(n))\n        counts = backend.run(transpile(qc, backend), shots=shots).result().get_counts()\n        for key, v in counts.items():\n            M[int(key, 2), i] = v \u002F shots\n    return M\n\nn = 2\nMinv = np.linalg.pinv(calibration_matrix(n, backend))\n\nbell = QuantumCircuit(2, 2)\nbell.h(0)\nbell.cx(0, 1)\nbell.measure([0, 1], [0, 1])\n\nraw = backend.run(transpile(bell, backend), shots=SHOTS).result().get_counts()\nvec = np.array([raw.get(format(i, f\"0{n}b\"), 0) for i in range(2**n)], float) \u002F SHOTS\n\nmit = np.clip(Minv @ vec, 0, None)\nmit \u002F= mit.sum()\n\nprint(\"raw      :\", {format(i, \"02b\"): round(vec[i], 4) for i in range(4)})\nprint(\"mitigated:\", {format(i, \"02b\"): round(mit[i], 4) for i in range(4)})\n",[4696],{"type":21,"tag":79,"props":4697,"children":4698},{"__ignoreMap":7},[4699,4721,4742,4763,4784,4791,4852,4869,4922,4965,4973,5008,5016,5029,5036,5066,5085,5093,5124,5169,5208,5226,5300,5328,5357,5366,5394,5425,5447,5493,5506,5514,5532,5550,5558,5592,5610,5635,5677,5685,5720,5836,5844,5889,5907,5915,6003],{"type":21,"tag":322,"props":4700,"children":4701},{"class":324,"line":325},[4702,4706,4711,4716],{"type":21,"tag":322,"props":4703,"children":4704},{"style":3638},[4705],{"type":26,"value":3651},{"type":21,"tag":322,"props":4707,"children":4708},{"style":329},[4709],{"type":26,"value":4710}," numpy ",{"type":21,"tag":322,"props":4712,"children":4713},{"style":3638},[4714],{"type":26,"value":4715},"as",{"type":21,"tag":322,"props":4717,"children":4718},{"style":329},[4719],{"type":26,"value":4720}," np\n",{"type":21,"tag":322,"props":4722,"children":4723},{"class":324,"line":335},[4724,4728,4733,4737],{"type":21,"tag":322,"props":4725,"children":4726},{"style":3638},[4727],{"type":26,"value":3641},{"type":21,"tag":322,"props":4729,"children":4730},{"style":329},[4731],{"type":26,"value":4732}," qiskit ",{"type":21,"tag":322,"props":4734,"children":4735},{"style":3638},[4736],{"type":26,"value":3651},{"type":21,"tag":322,"props":4738,"children":4739},{"style":329},[4740],{"type":26,"value":4741}," QuantumCircuit, transpile\n",{"type":21,"tag":322,"props":4743,"children":4744},{"class":324,"line":344},[4745,4749,4754,4758],{"type":21,"tag":322,"props":4746,"children":4747},{"style":3638},[4748],{"type":26,"value":3641},{"type":21,"tag":322,"props":4750,"children":4751},{"style":329},[4752],{"type":26,"value":4753}," qiskit_aer ",{"type":21,"tag":322,"props":4755,"children":4756},{"style":3638},[4757],{"type":26,"value":3651},{"type":21,"tag":322,"props":4759,"children":4760},{"style":329},[4761],{"type":26,"value":4762}," AerSimulator\n",{"type":21,"tag":322,"props":4764,"children":4765},{"class":324,"line":407},[4766,4770,4775,4779],{"type":21,"tag":322,"props":4767,"children":4768},{"style":3638},[4769],{"type":26,"value":3641},{"type":21,"tag":322,"props":4771,"children":4772},{"style":329},[4773],{"type":26,"value":4774}," qiskit_aer.noise ",{"type":21,"tag":322,"props":4776,"children":4777},{"style":3638},[4778],{"type":26,"value":3651},{"type":21,"tag":322,"props":4780,"children":4781},{"style":329},[4782],{"type":26,"value":4783}," NoiseModel, depolarizing_error, ReadoutError\n",{"type":21,"tag":322,"props":4785,"children":4786},{"class":324,"line":431},[4787],{"type":21,"tag":322,"props":4788,"children":4789},{"emptyLinePlaceholder":3662},[4790],{"type":26,"value":3665},{"type":21,"tag":322,"props":4792,"children":4793},{"class":324,"line":453},[4794,4799,4805,4810,4814,4819,4824,4828,4833,4838,4842,4847],{"type":21,"tag":322,"props":4795,"children":4796},{"style":3638},[4797],{"type":26,"value":4798},"def",{"type":21,"tag":322,"props":4800,"children":4802},{"style":4801},"--shiki-default:#B392F0",[4803],{"type":26,"value":4804}," build_noise",{"type":21,"tag":322,"props":4806,"children":4807},{"style":329},[4808],{"type":26,"value":4809},"(p1",{"type":21,"tag":322,"props":4811,"children":4812},{"style":3638},[4813],{"type":26,"value":3678},{"type":21,"tag":322,"props":4815,"children":4816},{"style":348},[4817],{"type":26,"value":4818},"0.002",{"type":21,"tag":322,"props":4820,"children":4821},{"style":329},[4822],{"type":26,"value":4823},", p2",{"type":21,"tag":322,"props":4825,"children":4826},{"style":3638},[4827],{"type":26,"value":3678},{"type":21,"tag":322,"props":4829,"children":4830},{"style":348},[4831],{"type":26,"value":4832},"0.02",{"type":21,"tag":322,"props":4834,"children":4835},{"style":329},[4836],{"type":26,"value":4837},", p_read",{"type":21,"tag":322,"props":4839,"children":4840},{"style":3638},[4841],{"type":26,"value":3678},{"type":21,"tag":322,"props":4843,"children":4844},{"style":348},[4845],{"type":26,"value":4846},"0.05",{"type":21,"tag":322,"props":4848,"children":4849},{"style":329},[4850],{"type":26,"value":4851},"):\n",{"type":21,"tag":322,"props":4853,"children":4854},{"class":324,"line":474},[4855,4860,4864],{"type":21,"tag":322,"props":4856,"children":4857},{"style":329},[4858],{"type":26,"value":4859},"    nm ",{"type":21,"tag":322,"props":4861,"children":4862},{"style":3638},[4863],{"type":26,"value":3678},{"type":21,"tag":322,"props":4865,"children":4866},{"style":329},[4867],{"type":26,"value":4868}," NoiseModel()\n",{"type":21,"tag":322,"props":4870,"children":4871},{"class":324,"line":496},[4872,4877,4881,4886,4891,4895,4900,4904,4909,4913,4918],{"type":21,"tag":322,"props":4873,"children":4874},{"style":329},[4875],{"type":26,"value":4876},"    nm.add_all_qubit_quantum_error(depolarizing_error(p1, ",{"type":21,"tag":322,"props":4878,"children":4879},{"style":348},[4880],{"type":26,"value":3994},{"type":21,"tag":322,"props":4882,"children":4883},{"style":329},[4884],{"type":26,"value":4885},"), [",{"type":21,"tag":322,"props":4887,"children":4888},{"style":359},[4889],{"type":26,"value":4890},"\"rz\"",{"type":21,"tag":322,"props":4892,"children":4893},{"style":329},[4894],{"type":26,"value":367},{"type":21,"tag":322,"props":4896,"children":4897},{"style":359},[4898],{"type":26,"value":4899},"\"sx\"",{"type":21,"tag":322,"props":4901,"children":4902},{"style":329},[4903],{"type":26,"value":367},{"type":21,"tag":322,"props":4905,"children":4906},{"style":359},[4907],{"type":26,"value":4908},"\"x\"",{"type":21,"tag":322,"props":4910,"children":4911},{"style":329},[4912],{"type":26,"value":367},{"type":21,"tag":322,"props":4914,"children":4915},{"style":359},[4916],{"type":26,"value":4917},"\"h\"",{"type":21,"tag":322,"props":4919,"children":4920},{"style":329},[4921],{"type":26,"value":3999},{"type":21,"tag":322,"props":4923,"children":4924},{"class":324,"line":16},[4925,4930,4934,4938,4943,4947,4952,4956,4961],{"type":21,"tag":322,"props":4926,"children":4927},{"style":329},[4928],{"type":26,"value":4929},"    nm.add_all_qubit_quantum_error(depolarizing_error(p2, ",{"type":21,"tag":322,"props":4931,"children":4932},{"style":348},[4933],{"type":26,"value":423},{"type":21,"tag":322,"props":4935,"children":4936},{"style":329},[4937],{"type":26,"value":4885},{"type":21,"tag":322,"props":4939,"children":4940},{"style":359},[4941],{"type":26,"value":4942},"\"cx\"",{"type":21,"tag":322,"props":4944,"children":4945},{"style":329},[4946],{"type":26,"value":367},{"type":21,"tag":322,"props":4948,"children":4949},{"style":359},[4950],{"type":26,"value":4951},"\"cz\"",{"type":21,"tag":322,"props":4953,"children":4954},{"style":329},[4955],{"type":26,"value":367},{"type":21,"tag":322,"props":4957,"children":4958},{"style":359},[4959],{"type":26,"value":4960},"\"ecr\"",{"type":21,"tag":322,"props":4962,"children":4963},{"style":329},[4964],{"type":26,"value":3999},{"type":21,"tag":322,"props":4966,"children":4967},{"class":324,"line":3884},[4968],{"type":21,"tag":322,"props":4969,"children":4970},{"style":329},[4971],{"type":26,"value":4972},"    nm.add_all_qubit_readout_error(\n",{"type":21,"tag":322,"props":4974,"children":4975},{"class":324,"line":3894},[4976,4981,4985,4990,4995,4999,5003],{"type":21,"tag":322,"props":4977,"children":4978},{"style":329},[4979],{"type":26,"value":4980},"        ReadoutError([[",{"type":21,"tag":322,"props":4982,"children":4983},{"style":348},[4984],{"type":26,"value":3994},{"type":21,"tag":322,"props":4986,"children":4987},{"style":3638},[4988],{"type":26,"value":4989}," -",{"type":21,"tag":322,"props":4991,"children":4992},{"style":329},[4993],{"type":26,"value":4994}," p_read, p_read], [p_read, ",{"type":21,"tag":322,"props":4996,"children":4997},{"style":348},[4998],{"type":26,"value":3994},{"type":21,"tag":322,"props":5000,"children":5001},{"style":3638},[5002],{"type":26,"value":4989},{"type":21,"tag":322,"props":5004,"children":5005},{"style":329},[5006],{"type":26,"value":5007}," p_read]])\n",{"type":21,"tag":322,"props":5009,"children":5010},{"class":324,"line":1034},[5011],{"type":21,"tag":322,"props":5012,"children":5013},{"style":329},[5014],{"type":26,"value":5015},"    )\n",{"type":21,"tag":322,"props":5017,"children":5018},{"class":324,"line":3954},[5019,5024],{"type":21,"tag":322,"props":5020,"children":5021},{"style":3638},[5022],{"type":26,"value":5023},"    return",{"type":21,"tag":322,"props":5025,"children":5026},{"style":329},[5027],{"type":26,"value":5028}," nm\n",{"type":21,"tag":322,"props":5030,"children":5031},{"class":324,"line":4002},[5032],{"type":21,"tag":322,"props":5033,"children":5034},{"emptyLinePlaceholder":3662},[5035],{"type":26,"value":3665},{"type":21,"tag":322,"props":5037,"children":5038},{"class":324,"line":4044},[5039,5043,5047,5052,5057,5061],{"type":21,"tag":322,"props":5040,"children":5041},{"style":329},[5042],{"type":26,"value":3691},{"type":21,"tag":322,"props":5044,"children":5045},{"style":3638},[5046],{"type":26,"value":3678},{"type":21,"tag":322,"props":5048,"children":5049},{"style":329},[5050],{"type":26,"value":5051}," AerSimulator(",{"type":21,"tag":322,"props":5053,"children":5054},{"style":3703},[5055],{"type":26,"value":5056},"noise_model",{"type":21,"tag":322,"props":5058,"children":5059},{"style":3638},[5060],{"type":26,"value":3678},{"type":21,"tag":322,"props":5062,"children":5063},{"style":329},[5064],{"type":26,"value":5065},"build_noise())\n",{"type":21,"tag":322,"props":5067,"children":5069},{"class":324,"line":5068},16,[5070,5075,5080],{"type":21,"tag":322,"props":5071,"children":5072},{"style":348},[5073],{"type":26,"value":5074},"SHOTS",{"type":21,"tag":322,"props":5076,"children":5077},{"style":3638},[5078],{"type":26,"value":5079}," =",{"type":21,"tag":322,"props":5081,"children":5082},{"style":348},[5083],{"type":26,"value":5084}," 40000\n",{"type":21,"tag":322,"props":5086,"children":5088},{"class":324,"line":5087},17,[5089],{"type":21,"tag":322,"props":5090,"children":5091},{"emptyLinePlaceholder":3662},[5092],{"type":26,"value":3665},{"type":21,"tag":322,"props":5094,"children":5096},{"class":324,"line":5095},18,[5097,5101,5106,5111,5115,5120],{"type":21,"tag":322,"props":5098,"children":5099},{"style":3638},[5100],{"type":26,"value":4798},{"type":21,"tag":322,"props":5102,"children":5103},{"style":4801},[5104],{"type":26,"value":5105}," calibration_matrix",{"type":21,"tag":322,"props":5107,"children":5108},{"style":329},[5109],{"type":26,"value":5110},"(n, backend, shots",{"type":21,"tag":322,"props":5112,"children":5113},{"style":3638},[5114],{"type":26,"value":3678},{"type":21,"tag":322,"props":5116,"children":5117},{"style":348},[5118],{"type":26,"value":5119},"20000",{"type":21,"tag":322,"props":5121,"children":5122},{"style":329},[5123],{"type":26,"value":4851},{"type":21,"tag":322,"props":5125,"children":5127},{"class":324,"line":5126},19,[5128,5133,5137,5142,5146,5151,5156,5160,5164],{"type":21,"tag":322,"props":5129,"children":5130},{"style":329},[5131],{"type":26,"value":5132},"    M ",{"type":21,"tag":322,"props":5134,"children":5135},{"style":3638},[5136],{"type":26,"value":3678},{"type":21,"tag":322,"props":5138,"children":5139},{"style":329},[5140],{"type":26,"value":5141}," np.zeros((",{"type":21,"tag":322,"props":5143,"children":5144},{"style":348},[5145],{"type":26,"value":423},{"type":21,"tag":322,"props":5147,"children":5148},{"style":3638},[5149],{"type":26,"value":5150},"**",{"type":21,"tag":322,"props":5152,"children":5153},{"style":329},[5154],{"type":26,"value":5155},"n, ",{"type":21,"tag":322,"props":5157,"children":5158},{"style":348},[5159],{"type":26,"value":423},{"type":21,"tag":322,"props":5161,"children":5162},{"style":3638},[5163],{"type":26,"value":5150},{"type":21,"tag":322,"props":5165,"children":5166},{"style":329},[5167],{"type":26,"value":5168},"n))\n",{"type":21,"tag":322,"props":5170,"children":5172},{"class":324,"line":5171},20,[5173,5178,5183,5187,5191,5195,5199,5203],{"type":21,"tag":322,"props":5174,"children":5175},{"style":3638},[5176],{"type":26,"value":5177},"    for",{"type":21,"tag":322,"props":5179,"children":5180},{"style":329},[5181],{"type":26,"value":5182}," i ",{"type":21,"tag":322,"props":5184,"children":5185},{"style":3638},[5186],{"type":26,"value":3941},{"type":21,"tag":322,"props":5188,"children":5189},{"style":348},[5190],{"type":26,"value":3946},{"type":21,"tag":322,"props":5192,"children":5193},{"style":329},[5194],{"type":26,"value":3758},{"type":21,"tag":322,"props":5196,"children":5197},{"style":348},[5198],{"type":26,"value":423},{"type":21,"tag":322,"props":5200,"children":5201},{"style":3638},[5202],{"type":26,"value":5150},{"type":21,"tag":322,"props":5204,"children":5205},{"style":329},[5206],{"type":26,"value":5207},"n):\n",{"type":21,"tag":322,"props":5209,"children":5211},{"class":324,"line":5210},21,[5212,5217,5221],{"type":21,"tag":322,"props":5213,"children":5214},{"style":329},[5215],{"type":26,"value":5216},"        qc ",{"type":21,"tag":322,"props":5218,"children":5219},{"style":3638},[5220],{"type":26,"value":3678},{"type":21,"tag":322,"props":5222,"children":5223},{"style":329},[5224],{"type":26,"value":5225}," QuantumCircuit(n, n)\n",{"type":21,"tag":322,"props":5227,"children":5229},{"class":324,"line":5228},22,[5230,5235,5239,5244,5249,5253,5258,5262,5267,5271,5276,5281,5286,5290,5295],{"type":21,"tag":322,"props":5231,"children":5232},{"style":329},[5233],{"type":26,"value":5234},"        bits ",{"type":21,"tag":322,"props":5236,"children":5237},{"style":3638},[5238],{"type":26,"value":3678},{"type":21,"tag":322,"props":5240,"children":5241},{"style":348},[5242],{"type":26,"value":5243}," format",{"type":21,"tag":322,"props":5245,"children":5246},{"style":329},[5247],{"type":26,"value":5248},"(i, ",{"type":21,"tag":322,"props":5250,"children":5251},{"style":3638},[5252],{"type":26,"value":3763},{"type":21,"tag":322,"props":5254,"children":5255},{"style":359},[5256],{"type":26,"value":5257},"\"0",{"type":21,"tag":322,"props":5259,"children":5260},{"style":348},[5261],{"type":26,"value":3773},{"type":21,"tag":322,"props":5263,"children":5264},{"style":329},[5265],{"type":26,"value":5266},"n",{"type":21,"tag":322,"props":5268,"children":5269},{"style":348},[5270],{"type":26,"value":3783},{"type":21,"tag":322,"props":5272,"children":5273},{"style":359},[5274],{"type":26,"value":5275},"b\"",{"type":21,"tag":322,"props":5277,"children":5278},{"style":329},[5279],{"type":26,"value":5280},")[::",{"type":21,"tag":322,"props":5282,"children":5283},{"style":3638},[5284],{"type":26,"value":5285},"-",{"type":21,"tag":322,"props":5287,"children":5288},{"style":348},[5289],{"type":26,"value":3994},{"type":21,"tag":322,"props":5291,"children":5292},{"style":329},[5293],{"type":26,"value":5294},"]      ",{"type":21,"tag":322,"props":5296,"children":5297},{"style":3888},[5298],{"type":26,"value":5299},"# Qiskit is little-endian\n",{"type":21,"tag":322,"props":5301,"children":5303},{"class":324,"line":5302},23,[5304,5309,5314,5318,5323],{"type":21,"tag":322,"props":5305,"children":5306},{"style":3638},[5307],{"type":26,"value":5308},"        for",{"type":21,"tag":322,"props":5310,"children":5311},{"style":329},[5312],{"type":26,"value":5313}," q, b ",{"type":21,"tag":322,"props":5315,"children":5316},{"style":3638},[5317],{"type":26,"value":3941},{"type":21,"tag":322,"props":5319,"children":5320},{"style":348},[5321],{"type":26,"value":5322}," enumerate",{"type":21,"tag":322,"props":5324,"children":5325},{"style":329},[5326],{"type":26,"value":5327},"(bits):\n",{"type":21,"tag":322,"props":5329,"children":5331},{"class":324,"line":5330},24,[5332,5337,5342,5347,5352],{"type":21,"tag":322,"props":5333,"children":5334},{"style":3638},[5335],{"type":26,"value":5336},"            if",{"type":21,"tag":322,"props":5338,"children":5339},{"style":329},[5340],{"type":26,"value":5341}," b ",{"type":21,"tag":322,"props":5343,"children":5344},{"style":3638},[5345],{"type":26,"value":5346},"==",{"type":21,"tag":322,"props":5348,"children":5349},{"style":359},[5350],{"type":26,"value":5351}," \"1\"",{"type":21,"tag":322,"props":5353,"children":5354},{"style":329},[5355],{"type":26,"value":5356},":\n",{"type":21,"tag":322,"props":5358,"children":5360},{"class":324,"line":5359},25,[5361],{"type":21,"tag":322,"props":5362,"children":5363},{"style":329},[5364],{"type":26,"value":5365},"                qc.x(q)\n",{"type":21,"tag":322,"props":5367,"children":5369},{"class":324,"line":5368},26,[5370,5375,5380,5385,5389],{"type":21,"tag":322,"props":5371,"children":5372},{"style":329},[5373],{"type":26,"value":5374},"        qc.measure(",{"type":21,"tag":322,"props":5376,"children":5377},{"style":348},[5378],{"type":26,"value":5379},"range",{"type":21,"tag":322,"props":5381,"children":5382},{"style":329},[5383],{"type":26,"value":5384},"(n), ",{"type":21,"tag":322,"props":5386,"children":5387},{"style":348},[5388],{"type":26,"value":5379},{"type":21,"tag":322,"props":5390,"children":5391},{"style":329},[5392],{"type":26,"value":5393},"(n))\n",{"type":21,"tag":322,"props":5395,"children":5397},{"class":324,"line":5396},27,[5398,5403,5407,5412,5416,5420],{"type":21,"tag":322,"props":5399,"children":5400},{"style":329},[5401],{"type":26,"value":5402},"        counts ",{"type":21,"tag":322,"props":5404,"children":5405},{"style":3638},[5406],{"type":26,"value":3678},{"type":21,"tag":322,"props":5408,"children":5409},{"style":329},[5410],{"type":26,"value":5411}," backend.run(transpile(qc, backend), ",{"type":21,"tag":322,"props":5413,"children":5414},{"style":3703},[5415],{"type":26,"value":4560},{"type":21,"tag":322,"props":5417,"children":5418},{"style":3638},[5419],{"type":26,"value":3678},{"type":21,"tag":322,"props":5421,"children":5422},{"style":329},[5423],{"type":26,"value":5424},"shots).result().get_counts()\n",{"type":21,"tag":322,"props":5426,"children":5428},{"class":324,"line":5427},28,[5429,5433,5438,5442],{"type":21,"tag":322,"props":5430,"children":5431},{"style":3638},[5432],{"type":26,"value":5308},{"type":21,"tag":322,"props":5434,"children":5435},{"style":329},[5436],{"type":26,"value":5437}," key, v ",{"type":21,"tag":322,"props":5439,"children":5440},{"style":3638},[5441],{"type":26,"value":3941},{"type":21,"tag":322,"props":5443,"children":5444},{"style":329},[5445],{"type":26,"value":5446}," counts.items():\n",{"type":21,"tag":322,"props":5448,"children":5450},{"class":324,"line":5449},29,[5451,5456,5461,5466,5470,5475,5479,5484,5488],{"type":21,"tag":322,"props":5452,"children":5453},{"style":329},[5454],{"type":26,"value":5455},"            M[",{"type":21,"tag":322,"props":5457,"children":5458},{"style":348},[5459],{"type":26,"value":5460},"int",{"type":21,"tag":322,"props":5462,"children":5463},{"style":329},[5464],{"type":26,"value":5465},"(key, ",{"type":21,"tag":322,"props":5467,"children":5468},{"style":348},[5469],{"type":26,"value":423},{"type":21,"tag":322,"props":5471,"children":5472},{"style":329},[5473],{"type":26,"value":5474},"), i] ",{"type":21,"tag":322,"props":5476,"children":5477},{"style":3638},[5478],{"type":26,"value":3678},{"type":21,"tag":322,"props":5480,"children":5481},{"style":329},[5482],{"type":26,"value":5483}," v ",{"type":21,"tag":322,"props":5485,"children":5486},{"style":3638},[5487],{"type":26,"value":174},{"type":21,"tag":322,"props":5489,"children":5490},{"style":329},[5491],{"type":26,"value":5492}," shots\n",{"type":21,"tag":322,"props":5494,"children":5496},{"class":324,"line":5495},30,[5497,5501],{"type":21,"tag":322,"props":5498,"children":5499},{"style":3638},[5500],{"type":26,"value":5023},{"type":21,"tag":322,"props":5502,"children":5503},{"style":329},[5504],{"type":26,"value":5505}," M\n",{"type":21,"tag":322,"props":5507,"children":5509},{"class":324,"line":5508},31,[5510],{"type":21,"tag":322,"props":5511,"children":5512},{"emptyLinePlaceholder":3662},[5513],{"type":26,"value":3665},{"type":21,"tag":322,"props":5515,"children":5517},{"class":324,"line":5516},32,[5518,5523,5527],{"type":21,"tag":322,"props":5519,"children":5520},{"style":329},[5521],{"type":26,"value":5522},"n ",{"type":21,"tag":322,"props":5524,"children":5525},{"style":3638},[5526],{"type":26,"value":3678},{"type":21,"tag":322,"props":5528,"children":5529},{"style":348},[5530],{"type":26,"value":5531}," 2\n",{"type":21,"tag":322,"props":5533,"children":5535},{"class":324,"line":5534},33,[5536,5541,5545],{"type":21,"tag":322,"props":5537,"children":5538},{"style":329},[5539],{"type":26,"value":5540},"Minv ",{"type":21,"tag":322,"props":5542,"children":5543},{"style":3638},[5544],{"type":26,"value":3678},{"type":21,"tag":322,"props":5546,"children":5547},{"style":329},[5548],{"type":26,"value":5549}," np.linalg.pinv(calibration_matrix(n, backend))\n",{"type":21,"tag":322,"props":5551,"children":5553},{"class":324,"line":5552},34,[5554],{"type":21,"tag":322,"props":5555,"children":5556},{"emptyLinePlaceholder":3662},[5557],{"type":26,"value":3665},{"type":21,"tag":322,"props":5559,"children":5561},{"class":324,"line":5560},35,[5562,5567,5571,5576,5580,5584,5588],{"type":21,"tag":322,"props":5563,"children":5564},{"style":329},[5565],{"type":26,"value":5566},"bell ",{"type":21,"tag":322,"props":5568,"children":5569},{"style":3638},[5570],{"type":26,"value":3678},{"type":21,"tag":322,"props":5572,"children":5573},{"style":329},[5574],{"type":26,"value":5575}," QuantumCircuit(",{"type":21,"tag":322,"props":5577,"children":5578},{"style":348},[5579],{"type":26,"value":423},{"type":21,"tag":322,"props":5581,"children":5582},{"style":329},[5583],{"type":26,"value":367},{"type":21,"tag":322,"props":5585,"children":5586},{"style":348},[5587],{"type":26,"value":423},{"type":21,"tag":322,"props":5589,"children":5590},{"style":329},[5591],{"type":26,"value":3738},{"type":21,"tag":322,"props":5593,"children":5595},{"class":324,"line":5594},36,[5596,5601,5606],{"type":21,"tag":322,"props":5597,"children":5598},{"style":329},[5599],{"type":26,"value":5600},"bell.h(",{"type":21,"tag":322,"props":5602,"children":5603},{"style":348},[5604],{"type":26,"value":5605},"0",{"type":21,"tag":322,"props":5607,"children":5608},{"style":329},[5609],{"type":26,"value":3738},{"type":21,"tag":322,"props":5611,"children":5613},{"class":324,"line":5612},37,[5614,5619,5623,5627,5631],{"type":21,"tag":322,"props":5615,"children":5616},{"style":329},[5617],{"type":26,"value":5618},"bell.cx(",{"type":21,"tag":322,"props":5620,"children":5621},{"style":348},[5622],{"type":26,"value":5605},{"type":21,"tag":322,"props":5624,"children":5625},{"style":329},[5626],{"type":26,"value":367},{"type":21,"tag":322,"props":5628,"children":5629},{"style":348},[5630],{"type":26,"value":3994},{"type":21,"tag":322,"props":5632,"children":5633},{"style":329},[5634],{"type":26,"value":3738},{"type":21,"tag":322,"props":5636,"children":5638},{"class":324,"line":5637},38,[5639,5644,5648,5652,5656,5661,5665,5669,5673],{"type":21,"tag":322,"props":5640,"children":5641},{"style":329},[5642],{"type":26,"value":5643},"bell.measure([",{"type":21,"tag":322,"props":5645,"children":5646},{"style":348},[5647],{"type":26,"value":5605},{"type":21,"tag":322,"props":5649,"children":5650},{"style":329},[5651],{"type":26,"value":367},{"type":21,"tag":322,"props":5653,"children":5654},{"style":348},[5655],{"type":26,"value":3994},{"type":21,"tag":322,"props":5657,"children":5658},{"style":329},[5659],{"type":26,"value":5660},"], [",{"type":21,"tag":322,"props":5662,"children":5663},{"style":348},[5664],{"type":26,"value":5605},{"type":21,"tag":322,"props":5666,"children":5667},{"style":329},[5668],{"type":26,"value":367},{"type":21,"tag":322,"props":5670,"children":5671},{"style":348},[5672],{"type":26,"value":3994},{"type":21,"tag":322,"props":5674,"children":5675},{"style":329},[5676],{"type":26,"value":3999},{"type":21,"tag":322,"props":5678,"children":5680},{"class":324,"line":5679},39,[5681],{"type":21,"tag":322,"props":5682,"children":5683},{"emptyLinePlaceholder":3662},[5684],{"type":26,"value":3665},{"type":21,"tag":322,"props":5686,"children":5688},{"class":324,"line":5687},40,[5689,5694,5698,5703,5707,5711,5715],{"type":21,"tag":322,"props":5690,"children":5691},{"style":329},[5692],{"type":26,"value":5693},"raw ",{"type":21,"tag":322,"props":5695,"children":5696},{"style":3638},[5697],{"type":26,"value":3678},{"type":21,"tag":322,"props":5699,"children":5700},{"style":329},[5701],{"type":26,"value":5702}," backend.run(transpile(bell, backend), ",{"type":21,"tag":322,"props":5704,"children":5705},{"style":3703},[5706],{"type":26,"value":4560},{"type":21,"tag":322,"props":5708,"children":5709},{"style":3638},[5710],{"type":26,"value":3678},{"type":21,"tag":322,"props":5712,"children":5713},{"style":348},[5714],{"type":26,"value":5074},{"type":21,"tag":322,"props":5716,"children":5717},{"style":329},[5718],{"type":26,"value":5719},").result().get_counts()\n",{"type":21,"tag":322,"props":5721,"children":5723},{"class":324,"line":5722},41,[5724,5729,5733,5738,5743,5747,5751,5755,5759,5763,5767,5771,5776,5780,5785,5789,5793,5797,5801,5805,5809,5813,5818,5823,5827,5831],{"type":21,"tag":322,"props":5725,"children":5726},{"style":329},[5727],{"type":26,"value":5728},"vec ",{"type":21,"tag":322,"props":5730,"children":5731},{"style":3638},[5732],{"type":26,"value":3678},{"type":21,"tag":322,"props":5734,"children":5735},{"style":329},[5736],{"type":26,"value":5737}," np.array([raw.get(",{"type":21,"tag":322,"props":5739,"children":5740},{"style":348},[5741],{"type":26,"value":5742},"format",{"type":21,"tag":322,"props":5744,"children":5745},{"style":329},[5746],{"type":26,"value":5248},{"type":21,"tag":322,"props":5748,"children":5749},{"style":3638},[5750],{"type":26,"value":3763},{"type":21,"tag":322,"props":5752,"children":5753},{"style":359},[5754],{"type":26,"value":5257},{"type":21,"tag":322,"props":5756,"children":5757},{"style":348},[5758],{"type":26,"value":3773},{"type":21,"tag":322,"props":5760,"children":5761},{"style":329},[5762],{"type":26,"value":5266},{"type":21,"tag":322,"props":5764,"children":5765},{"style":348},[5766],{"type":26,"value":3783},{"type":21,"tag":322,"props":5768,"children":5769},{"style":359},[5770],{"type":26,"value":5275},{"type":21,"tag":322,"props":5772,"children":5773},{"style":329},[5774],{"type":26,"value":5775},"), ",{"type":21,"tag":322,"props":5777,"children":5778},{"style":348},[5779],{"type":26,"value":5605},{"type":21,"tag":322,"props":5781,"children":5782},{"style":329},[5783],{"type":26,"value":5784},") ",{"type":21,"tag":322,"props":5786,"children":5787},{"style":3638},[5788],{"type":26,"value":3931},{"type":21,"tag":322,"props":5790,"children":5791},{"style":329},[5792],{"type":26,"value":5182},{"type":21,"tag":322,"props":5794,"children":5795},{"style":3638},[5796],{"type":26,"value":3941},{"type":21,"tag":322,"props":5798,"children":5799},{"style":348},[5800],{"type":26,"value":3946},{"type":21,"tag":322,"props":5802,"children":5803},{"style":329},[5804],{"type":26,"value":3758},{"type":21,"tag":322,"props":5806,"children":5807},{"style":348},[5808],{"type":26,"value":423},{"type":21,"tag":322,"props":5810,"children":5811},{"style":3638},[5812],{"type":26,"value":5150},{"type":21,"tag":322,"props":5814,"children":5815},{"style":329},[5816],{"type":26,"value":5817},"n)], ",{"type":21,"tag":322,"props":5819,"children":5820},{"style":348},[5821],{"type":26,"value":5822},"float",{"type":21,"tag":322,"props":5824,"children":5825},{"style":329},[5826],{"type":26,"value":5784},{"type":21,"tag":322,"props":5828,"children":5829},{"style":3638},[5830],{"type":26,"value":174},{"type":21,"tag":322,"props":5832,"children":5833},{"style":348},[5834],{"type":26,"value":5835}," SHOTS\n",{"type":21,"tag":322,"props":5837,"children":5839},{"class":324,"line":5838},42,[5840],{"type":21,"tag":322,"props":5841,"children":5842},{"emptyLinePlaceholder":3662},[5843],{"type":26,"value":3665},{"type":21,"tag":322,"props":5845,"children":5847},{"class":324,"line":5846},43,[5848,5853,5857,5862,5867,5872,5876,5880,5885],{"type":21,"tag":322,"props":5849,"children":5850},{"style":329},[5851],{"type":26,"value":5852},"mit ",{"type":21,"tag":322,"props":5854,"children":5855},{"style":3638},[5856],{"type":26,"value":3678},{"type":21,"tag":322,"props":5858,"children":5859},{"style":329},[5860],{"type":26,"value":5861}," np.clip(Minv ",{"type":21,"tag":322,"props":5863,"children":5864},{"style":3638},[5865],{"type":26,"value":5866},"@",{"type":21,"tag":322,"props":5868,"children":5869},{"style":329},[5870],{"type":26,"value":5871}," vec, ",{"type":21,"tag":322,"props":5873,"children":5874},{"style":348},[5875],{"type":26,"value":5605},{"type":21,"tag":322,"props":5877,"children":5878},{"style":329},[5879],{"type":26,"value":367},{"type":21,"tag":322,"props":5881,"children":5882},{"style":348},[5883],{"type":26,"value":5884},"None",{"type":21,"tag":322,"props":5886,"children":5887},{"style":329},[5888],{"type":26,"value":3738},{"type":21,"tag":322,"props":5890,"children":5892},{"class":324,"line":5891},44,[5893,5897,5902],{"type":21,"tag":322,"props":5894,"children":5895},{"style":329},[5896],{"type":26,"value":5852},{"type":21,"tag":322,"props":5898,"children":5899},{"style":3638},[5900],{"type":26,"value":5901},"\u002F=",{"type":21,"tag":322,"props":5903,"children":5904},{"style":329},[5905],{"type":26,"value":5906}," mit.sum()\n",{"type":21,"tag":322,"props":5908,"children":5910},{"class":324,"line":5909},45,[5911],{"type":21,"tag":322,"props":5912,"children":5913},{"emptyLinePlaceholder":3662},[5914],{"type":26,"value":3665},{"type":21,"tag":322,"props":5916,"children":5918},{"class":324,"line":5917},46,[5919,5923,5927,5932,5937,5941,5945,5950,5955,5960,5965,5970,5974,5978,5982,5986,5990,5994,5998],{"type":21,"tag":322,"props":5920,"children":5921},{"style":348},[5922],{"type":26,"value":3753},{"type":21,"tag":322,"props":5924,"children":5925},{"style":329},[5926],{"type":26,"value":3758},{"type":21,"tag":322,"props":5928,"children":5929},{"style":359},[5930],{"type":26,"value":5931},"\"raw      :\"",{"type":21,"tag":322,"props":5933,"children":5934},{"style":329},[5935],{"type":26,"value":5936},", {",{"type":21,"tag":322,"props":5938,"children":5939},{"style":348},[5940],{"type":26,"value":5742},{"type":21,"tag":322,"props":5942,"children":5943},{"style":329},[5944],{"type":26,"value":5248},{"type":21,"tag":322,"props":5946,"children":5947},{"style":359},[5948],{"type":26,"value":5949},"\"02b\"",{"type":21,"tag":322,"props":5951,"children":5952},{"style":329},[5953],{"type":26,"value":5954},"): ",{"type":21,"tag":322,"props":5956,"children":5957},{"style":348},[5958],{"type":26,"value":5959},"round",{"type":21,"tag":322,"props":5961,"children":5962},{"style":329},[5963],{"type":26,"value":5964},"(vec[i], ",{"type":21,"tag":322,"props":5966,"children":5967},{"style":348},[5968],{"type":26,"value":5969},"4",{"type":21,"tag":322,"props":5971,"children":5972},{"style":329},[5973],{"type":26,"value":5784},{"type":21,"tag":322,"props":5975,"children":5976},{"style":3638},[5977],{"type":26,"value":3931},{"type":21,"tag":322,"props":5979,"children":5980},{"style":329},[5981],{"type":26,"value":5182},{"type":21,"tag":322,"props":5983,"children":5984},{"style":3638},[5985],{"type":26,"value":3941},{"type":21,"tag":322,"props":5987,"children":5988},{"style":348},[5989],{"type":26,"value":3946},{"type":21,"tag":322,"props":5991,"children":5992},{"style":329},[5993],{"type":26,"value":3758},{"type":21,"tag":322,"props":5995,"children":5996},{"style":348},[5997],{"type":26,"value":5969},{"type":21,"tag":322,"props":5999,"children":6000},{"style":329},[6001],{"type":26,"value":6002},")})\n",{"type":21,"tag":322,"props":6004,"children":6006},{"class":324,"line":6005},47,[6007,6011,6015,6020,6024,6028,6032,6036,6040,6044,6049,6053,6057,6061,6065,6069,6073,6077,6081],{"type":21,"tag":322,"props":6008,"children":6009},{"style":348},[6010],{"type":26,"value":3753},{"type":21,"tag":322,"props":6012,"children":6013},{"style":329},[6014],{"type":26,"value":3758},{"type":21,"tag":322,"props":6016,"children":6017},{"style":359},[6018],{"type":26,"value":6019},"\"mitigated:\"",{"type":21,"tag":322,"props":6021,"children":6022},{"style":329},[6023],{"type":26,"value":5936},{"type":21,"tag":322,"props":6025,"children":6026},{"style":348},[6027],{"type":26,"value":5742},{"type":21,"tag":322,"props":6029,"children":6030},{"style":329},[6031],{"type":26,"value":5248},{"type":21,"tag":322,"props":6033,"children":6034},{"style":359},[6035],{"type":26,"value":5949},{"type":21,"tag":322,"props":6037,"children":6038},{"style":329},[6039],{"type":26,"value":5954},{"type":21,"tag":322,"props":6041,"children":6042},{"style":348},[6043],{"type":26,"value":5959},{"type":21,"tag":322,"props":6045,"children":6046},{"style":329},[6047],{"type":26,"value":6048},"(mit[i], ",{"type":21,"tag":322,"props":6050,"children":6051},{"style":348},[6052],{"type":26,"value":5969},{"type":21,"tag":322,"props":6054,"children":6055},{"style":329},[6056],{"type":26,"value":5784},{"type":21,"tag":322,"props":6058,"children":6059},{"style":3638},[6060],{"type":26,"value":3931},{"type":21,"tag":322,"props":6062,"children":6063},{"style":329},[6064],{"type":26,"value":5182},{"type":21,"tag":322,"props":6066,"children":6067},{"style":3638},[6068],{"type":26,"value":3941},{"type":21,"tag":322,"props":6070,"children":6071},{"style":348},[6072],{"type":26,"value":3946},{"type":21,"tag":322,"props":6074,"children":6075},{"style":329},[6076],{"type":26,"value":3758},{"type":21,"tag":322,"props":6078,"children":6079},{"style":348},[6080],{"type":26,"value":5969},{"type":21,"tag":322,"props":6082,"children":6083},{"style":329},[6084],{"type":26,"value":6002},{"type":21,"tag":22,"props":6086,"children":6087},{},[6088,6093],{"type":21,"tag":71,"props":6089,"children":6090},{},[6091],{"type":26,"value":6092},"Actual output from running this",{"type":26,"value":6094}," (Qiskit 2.5.0, Aer 0.17.2, 5% readout error):",{"type":21,"tag":312,"props":6096,"children":6098},{"code":6097},"raw      : {'00': 0.4456, '01': 0.053, '10': 0.0501, '11': 0.4513}\nmitigated: {'00': 0.4931, '01': 0.0048, '10': 0.0033, '11': 0.4988}\n",[6099],{"type":21,"tag":79,"props":6100,"children":6101},{"__ignoreMap":7},[6102],{"type":26,"value":6097},{"type":21,"tag":22,"props":6104,"children":6105},{},[6106,6108,6114,6115,6121,6123,6129],{"type":26,"value":6107},"A Bell state should produce only ",{"type":21,"tag":79,"props":6109,"children":6111},{"className":6110},[],[6112],{"type":26,"value":6113},"00",{"type":26,"value":236},{"type":21,"tag":79,"props":6116,"children":6118},{"className":6117},[],[6119],{"type":26,"value":6120},"11",{"type":26,"value":6122},". The raw run leaks 10.3% into forbidden outcomes. After mitigation, 0.8%. That is a 13× reduction in spurious counts for the cost of ",{"type":21,"tag":79,"props":6124,"children":6126},{"className":6125},[],[6127],{"type":26,"value":6128},"2^n",{"type":26,"value":6130}," short calibration circuits.",{"type":21,"tag":22,"props":6132,"children":6133},{},[6134,6136,6141,6143,6149,6151,6156],{"type":26,"value":6135},"Two caveats. The ",{"type":21,"tag":79,"props":6137,"children":6139},{"className":6138},[],[6140],{"type":26,"value":6128},{"type":26,"value":6142}," scaling means full calibration matrices die above ~10 qubits. Beyond that, use tensored (per-qubit) or correlated-subset calibration. And matrix inversion amplifies ",{"type":21,"tag":29,"props":6144,"children":6146},{"href":6145},"\u002Fglossary\u002Fshot-noise",[6147],{"type":26,"value":6148},"shot noise",{"type":26,"value":6150},", which is why the code clips negatives. Run your calibration circuits with ",{"type":21,"tag":51,"props":6152,"children":6153},{},[6154],{"type":26,"value":6155},"more",{"type":26,"value":6157}," shots than your actual experiment, not fewer.",{"type":21,"tag":59,"props":6159,"children":6161},{"id":6160},"step-2-dynamical-decoupling-nearly-free",[6162],{"type":26,"value":6163},"Step 2: Dynamical Decoupling (Nearly Free)",{"type":21,"tag":22,"props":6165,"children":6166},{},[6167,6169,6175,6177,6183],{"type":26,"value":6168},"Idle qubits are not safe qubits. While one part of your circuit executes, everything else is dephasing at its ",{"type":21,"tag":29,"props":6170,"children":6172},{"href":6171},"\u002Fglossary\u002Ft1-t2-time",[6173],{"type":26,"value":6174},"T2 rate",{"type":26,"value":6176},". Dynamical decoupling inserts pulse sequences (typically ",{"type":21,"tag":79,"props":6178,"children":6180},{"className":6179},[],[6181],{"type":26,"value":6182},"X–delay–X–delay",{"type":26,"value":6184},") into idle windows, echoing away slow, low-frequency noise.",{"type":21,"tag":22,"props":6186,"children":6187},{},[6188,6190,6196],{"type":26,"value":6189},"In Qiskit this is a ",{"type":21,"tag":29,"props":6191,"children":6193},{"href":6192},"\u002Fglossary\u002Ftranspilation",[6194],{"type":26,"value":6195},"transpilation",{"type":26,"value":6197}," pass, not an algorithm change:",{"type":21,"tag":312,"props":6199,"children":6201},{"code":6200,"language":3628,"meta":7,"className":3626,"style":7},"from qiskit.transpiler import PassManager\nfrom qiskit.transpiler.passes import ALAPScheduleAnalysis, PadDynamicalDecoupling\nfrom qiskit.circuit.library import XGate\n\npm = PassManager([\n    ALAPScheduleAnalysis(durations=backend.instruction_durations),\n    PadDynamicalDecoupling(\n        durations=backend.instruction_durations,\n        dd_sequence=[XGate(), XGate()],   # XX sequence\n    ),\n])\ndd_circuit = pm.run(transpiled_circuit)\n",[6202],{"type":21,"tag":79,"props":6203,"children":6204},{"__ignoreMap":7},[6205,6226,6247,6268,6275,6292,6314,6322,6339,6361,6369,6376],{"type":21,"tag":322,"props":6206,"children":6207},{"class":324,"line":325},[6208,6212,6217,6221],{"type":21,"tag":322,"props":6209,"children":6210},{"style":3638},[6211],{"type":26,"value":3641},{"type":21,"tag":322,"props":6213,"children":6214},{"style":329},[6215],{"type":26,"value":6216}," qiskit.transpiler ",{"type":21,"tag":322,"props":6218,"children":6219},{"style":3638},[6220],{"type":26,"value":3651},{"type":21,"tag":322,"props":6222,"children":6223},{"style":329},[6224],{"type":26,"value":6225}," PassManager\n",{"type":21,"tag":322,"props":6227,"children":6228},{"class":324,"line":335},[6229,6233,6238,6242],{"type":21,"tag":322,"props":6230,"children":6231},{"style":3638},[6232],{"type":26,"value":3641},{"type":21,"tag":322,"props":6234,"children":6235},{"style":329},[6236],{"type":26,"value":6237}," qiskit.transpiler.passes ",{"type":21,"tag":322,"props":6239,"children":6240},{"style":3638},[6241],{"type":26,"value":3651},{"type":21,"tag":322,"props":6243,"children":6244},{"style":329},[6245],{"type":26,"value":6246}," ALAPScheduleAnalysis, PadDynamicalDecoupling\n",{"type":21,"tag":322,"props":6248,"children":6249},{"class":324,"line":344},[6250,6254,6259,6263],{"type":21,"tag":322,"props":6251,"children":6252},{"style":3638},[6253],{"type":26,"value":3641},{"type":21,"tag":322,"props":6255,"children":6256},{"style":329},[6257],{"type":26,"value":6258}," qiskit.circuit.library ",{"type":21,"tag":322,"props":6260,"children":6261},{"style":3638},[6262],{"type":26,"value":3651},{"type":21,"tag":322,"props":6264,"children":6265},{"style":329},[6266],{"type":26,"value":6267}," XGate\n",{"type":21,"tag":322,"props":6269,"children":6270},{"class":324,"line":407},[6271],{"type":21,"tag":322,"props":6272,"children":6273},{"emptyLinePlaceholder":3662},[6274],{"type":26,"value":3665},{"type":21,"tag":322,"props":6276,"children":6277},{"class":324,"line":431},[6278,6283,6287],{"type":21,"tag":322,"props":6279,"children":6280},{"style":329},[6281],{"type":26,"value":6282},"pm ",{"type":21,"tag":322,"props":6284,"children":6285},{"style":3638},[6286],{"type":26,"value":3678},{"type":21,"tag":322,"props":6288,"children":6289},{"style":329},[6290],{"type":26,"value":6291}," PassManager([\n",{"type":21,"tag":322,"props":6293,"children":6294},{"class":324,"line":453},[6295,6300,6305,6309],{"type":21,"tag":322,"props":6296,"children":6297},{"style":329},[6298],{"type":26,"value":6299},"    ALAPScheduleAnalysis(",{"type":21,"tag":322,"props":6301,"children":6302},{"style":3703},[6303],{"type":26,"value":6304},"durations",{"type":21,"tag":322,"props":6306,"children":6307},{"style":3638},[6308],{"type":26,"value":3678},{"type":21,"tag":322,"props":6310,"children":6311},{"style":329},[6312],{"type":26,"value":6313},"backend.instruction_durations),\n",{"type":21,"tag":322,"props":6315,"children":6316},{"class":324,"line":474},[6317],{"type":21,"tag":322,"props":6318,"children":6319},{"style":329},[6320],{"type":26,"value":6321},"    PadDynamicalDecoupling(\n",{"type":21,"tag":322,"props":6323,"children":6324},{"class":324,"line":496},[6325,6330,6334],{"type":21,"tag":322,"props":6326,"children":6327},{"style":3703},[6328],{"type":26,"value":6329},"        durations",{"type":21,"tag":322,"props":6331,"children":6332},{"style":3638},[6333],{"type":26,"value":3678},{"type":21,"tag":322,"props":6335,"children":6336},{"style":329},[6337],{"type":26,"value":6338},"backend.instruction_durations,\n",{"type":21,"tag":322,"props":6340,"children":6341},{"class":324,"line":16},[6342,6347,6351,6356],{"type":21,"tag":322,"props":6343,"children":6344},{"style":3703},[6345],{"type":26,"value":6346},"        dd_sequence",{"type":21,"tag":322,"props":6348,"children":6349},{"style":3638},[6350],{"type":26,"value":3678},{"type":21,"tag":322,"props":6352,"children":6353},{"style":329},[6354],{"type":26,"value":6355},"[XGate(), XGate()],   ",{"type":21,"tag":322,"props":6357,"children":6358},{"style":3888},[6359],{"type":26,"value":6360},"# XX sequence\n",{"type":21,"tag":322,"props":6362,"children":6363},{"class":324,"line":3884},[6364],{"type":21,"tag":322,"props":6365,"children":6366},{"style":329},[6367],{"type":26,"value":6368},"    ),\n",{"type":21,"tag":322,"props":6370,"children":6371},{"class":324,"line":3894},[6372],{"type":21,"tag":322,"props":6373,"children":6374},{"style":329},[6375],{"type":26,"value":3999},{"type":21,"tag":322,"props":6377,"children":6378},{"class":324,"line":1034},[6379,6384,6388],{"type":21,"tag":322,"props":6380,"children":6381},{"style":329},[6382],{"type":26,"value":6383},"dd_circuit ",{"type":21,"tag":322,"props":6385,"children":6386},{"style":3638},[6387],{"type":26,"value":3678},{"type":21,"tag":322,"props":6389,"children":6390},{"style":329},[6391],{"type":26,"value":6392}," pm.run(transpiled_circuit)\n",{"type":21,"tag":22,"props":6394,"children":6395},{},[6396],{"type":26,"value":6397},"Cost: zero extra shots. Typical gain: 5–20% fidelity improvement on circuits with heterogeneous qubit usage. Circuits where every qubit is busy every cycle see nothing. Try it. It's the best effort-to-payoff ratio in the whole toolkit.",{"type":21,"tag":59,"props":6399,"children":6401},{"id":6400},"step-3-pauli-twirling-randomized-compiling",[6402],{"type":26,"value":6403},"Step 3: Pauli Twirling \u002F Randomized Compiling",{"type":21,"tag":22,"props":6405,"children":6406},{},[6407,6409,6414],{"type":26,"value":6408},"Coherent errors are the nasty kind. A systematic over-rotation of 0.5° per gate doesn't average out. It accumulates ",{"type":21,"tag":51,"props":6410,"children":6411},{},[6412],{"type":26,"value":6413},"linearly",{"type":26,"value":6415}," in amplitude, so error in the expectation value grows quadratically with depth. Stochastic errors merely accumulate additively.",{"type":21,"tag":22,"props":6417,"children":6418},{},[6419],{"type":26,"value":6420},"Twirling converts the former into the latter. You compile many logically equivalent copies of your circuit, each wrapping the two-qubit gates in randomly chosen Pauli operators (with compensating Paulis so the ideal unitary is unchanged), then average the results. Coherent error terms average toward zero. What survives is a well-behaved stochastic Pauli channel.",{"type":21,"tag":22,"props":6422,"children":6423},{},[6424,6426,6431],{"type":26,"value":6425},"Twirling rarely improves accuracy much on its own. Its real value is that it makes the noise ",{"type":21,"tag":51,"props":6427,"children":6428},{},[6429],{"type":26,"value":6430},"model-able",{"type":26,"value":6432},", and every other technique on this list (ZNE, PEC) assumes a stochastic noise channel. Twirling is what makes that assumption true. Run it underneath your other mitigation, not instead of it.",{"type":21,"tag":59,"props":6434,"children":6436},{"id":6435},"step-4-zero-noise-extrapolation",[6437],{"type":26,"value":6438},"Step 4: Zero-Noise Extrapolation",{"type":21,"tag":22,"props":6440,"children":6441},{},[6442,6444,6449],{"type":26,"value":6443},"ZNE is the technique behind most headline NISQ results, including IBM's ",{"type":21,"tag":29,"props":6445,"children":6446},{"href":2054},[6447],{"type":26,"value":6448},"quantum utility demonstration",{"type":26,"value":6450}," on a 127-qubit Eagle processor.",{"type":21,"tag":22,"props":6452,"children":6453},{},[6454,6456,6461,6463,6469,6471,6477],{"type":26,"value":6455},"The idea: you can't reduce hardware noise, but you ",{"type":21,"tag":51,"props":6457,"children":6458},{},[6459],{"type":26,"value":6460},"increase",{"type":26,"value":6462}," it in a controlled way. Run the same circuit at noise scale factors λ = 1, 3, 5 (via unitary folding: replace ",{"type":21,"tag":79,"props":6464,"children":6466},{"className":6465},[],[6467],{"type":26,"value":6468},"U",{"type":26,"value":6470}," with ",{"type":21,"tag":79,"props":6472,"children":6474},{"className":6473},[],[6475],{"type":26,"value":6476},"U(U†U)^k",{"type":26,"value":6478},", which is logically identity-preserving but physically 3× or 5× as noisy), then fit a curve through the results and extrapolate back to λ = 0.",{"type":21,"tag":312,"props":6480,"children":6482},{"code":6481,"language":3628,"meta":7,"className":3626,"style":7},"def fold_global(qc, scale):\n    \"\"\"Odd integer scale: U -> U (U^dag U)^k\"\"\"\n    k = (scale - 1) \u002F\u002F 2\n    out, inv = qc.copy(), qc.inverse()\n    for _ in range(k):\n        out = out.compose(inv).compose(qc)\n    return out\n\ndef expval_parity(qc, backend, shots=40000):\n    c = qc.copy()\n    c.measure_all()\n    counts = backend.run(\n        transpile(c, backend, optimization_level=0), shots=shots\n    ).result().get_counts()\n    return sum((-1) ** (k.replace(\" \", \"\").count(\"1\") % 2) * v\n               for k, v in counts.items()) \u002F shots\n\n# 4-qubit GHZ padded with cancelling CX pairs, ideal \u003CZZZZ> = +1.0\nbase = QuantumCircuit(4)\nbase.h(0)\nfor q in range(3):\n    base.cx(q, q + 1)\nfor _ in range(4):\n    for q in range(3):\n        base.cx(q, q + 1)\n        base.cx(q, q + 1)\n\nscales = [1, 3, 5]\nvals = [expval_parity(fold_global(base, s), backend) for s in scales]\nzne = np.polyval(np.polyfit(scales, vals, 2), 0)   # Richardson, order 2\nprint(vals, \"->\", zne)\n",[6483],{"type":21,"tag":79,"props":6484,"children":6485},{"__ignoreMap":7},[6486,6503,6511,6550,6567,6592,6609,6621,6628,6658,6675,6683,6700,6738,6746,6836,6866,6873,6881,6905,6921,6953,6974,7005,7036,7056,7075,7082,7125,7160,7199],{"type":21,"tag":322,"props":6487,"children":6488},{"class":324,"line":325},[6489,6493,6498],{"type":21,"tag":322,"props":6490,"children":6491},{"style":3638},[6492],{"type":26,"value":4798},{"type":21,"tag":322,"props":6494,"children":6495},{"style":4801},[6496],{"type":26,"value":6497}," fold_global",{"type":21,"tag":322,"props":6499,"children":6500},{"style":329},[6501],{"type":26,"value":6502},"(qc, scale):\n",{"type":21,"tag":322,"props":6504,"children":6505},{"class":324,"line":335},[6506],{"type":21,"tag":322,"props":6507,"children":6508},{"style":359},[6509],{"type":26,"value":6510},"    \"\"\"Odd integer scale: U -> U (U^dag U)^k\"\"\"\n",{"type":21,"tag":322,"props":6512,"children":6513},{"class":324,"line":344},[6514,6519,6523,6528,6532,6537,6541,6546],{"type":21,"tag":322,"props":6515,"children":6516},{"style":329},[6517],{"type":26,"value":6518},"    k ",{"type":21,"tag":322,"props":6520,"children":6521},{"style":3638},[6522],{"type":26,"value":3678},{"type":21,"tag":322,"props":6524,"children":6525},{"style":329},[6526],{"type":26,"value":6527}," (scale ",{"type":21,"tag":322,"props":6529,"children":6530},{"style":3638},[6531],{"type":26,"value":5285},{"type":21,"tag":322,"props":6533,"children":6534},{"style":348},[6535],{"type":26,"value":6536}," 1",{"type":21,"tag":322,"props":6538,"children":6539},{"style":329},[6540],{"type":26,"value":5784},{"type":21,"tag":322,"props":6542,"children":6543},{"style":3638},[6544],{"type":26,"value":6545},"\u002F\u002F",{"type":21,"tag":322,"props":6547,"children":6548},{"style":348},[6549],{"type":26,"value":5531},{"type":21,"tag":322,"props":6551,"children":6552},{"class":324,"line":407},[6553,6558,6562],{"type":21,"tag":322,"props":6554,"children":6555},{"style":329},[6556],{"type":26,"value":6557},"    out, inv ",{"type":21,"tag":322,"props":6559,"children":6560},{"style":3638},[6561],{"type":26,"value":3678},{"type":21,"tag":322,"props":6563,"children":6564},{"style":329},[6565],{"type":26,"value":6566}," qc.copy(), qc.inverse()\n",{"type":21,"tag":322,"props":6568,"children":6569},{"class":324,"line":431},[6570,6574,6579,6583,6587],{"type":21,"tag":322,"props":6571,"children":6572},{"style":3638},[6573],{"type":26,"value":5177},{"type":21,"tag":322,"props":6575,"children":6576},{"style":329},[6577],{"type":26,"value":6578}," _ ",{"type":21,"tag":322,"props":6580,"children":6581},{"style":3638},[6582],{"type":26,"value":3941},{"type":21,"tag":322,"props":6584,"children":6585},{"style":348},[6586],{"type":26,"value":3946},{"type":21,"tag":322,"props":6588,"children":6589},{"style":329},[6590],{"type":26,"value":6591},"(k):\n",{"type":21,"tag":322,"props":6593,"children":6594},{"class":324,"line":453},[6595,6600,6604],{"type":21,"tag":322,"props":6596,"children":6597},{"style":329},[6598],{"type":26,"value":6599},"        out ",{"type":21,"tag":322,"props":6601,"children":6602},{"style":3638},[6603],{"type":26,"value":3678},{"type":21,"tag":322,"props":6605,"children":6606},{"style":329},[6607],{"type":26,"value":6608}," out.compose(inv).compose(qc)\n",{"type":21,"tag":322,"props":6610,"children":6611},{"class":324,"line":474},[6612,6616],{"type":21,"tag":322,"props":6613,"children":6614},{"style":3638},[6615],{"type":26,"value":5023},{"type":21,"tag":322,"props":6617,"children":6618},{"style":329},[6619],{"type":26,"value":6620}," out\n",{"type":21,"tag":322,"props":6622,"children":6623},{"class":324,"line":496},[6624],{"type":21,"tag":322,"props":6625,"children":6626},{"emptyLinePlaceholder":3662},[6627],{"type":26,"value":3665},{"type":21,"tag":322,"props":6629,"children":6630},{"class":324,"line":16},[6631,6635,6640,6645,6649,6654],{"type":21,"tag":322,"props":6632,"children":6633},{"style":3638},[6634],{"type":26,"value":4798},{"type":21,"tag":322,"props":6636,"children":6637},{"style":4801},[6638],{"type":26,"value":6639}," expval_parity",{"type":21,"tag":322,"props":6641,"children":6642},{"style":329},[6643],{"type":26,"value":6644},"(qc, backend, shots",{"type":21,"tag":322,"props":6646,"children":6647},{"style":3638},[6648],{"type":26,"value":3678},{"type":21,"tag":322,"props":6650,"children":6651},{"style":348},[6652],{"type":26,"value":6653},"40000",{"type":21,"tag":322,"props":6655,"children":6656},{"style":329},[6657],{"type":26,"value":4851},{"type":21,"tag":322,"props":6659,"children":6660},{"class":324,"line":3884},[6661,6666,6670],{"type":21,"tag":322,"props":6662,"children":6663},{"style":329},[6664],{"type":26,"value":6665},"    c ",{"type":21,"tag":322,"props":6667,"children":6668},{"style":3638},[6669],{"type":26,"value":3678},{"type":21,"tag":322,"props":6671,"children":6672},{"style":329},[6673],{"type":26,"value":6674}," qc.copy()\n",{"type":21,"tag":322,"props":6676,"children":6677},{"class":324,"line":3894},[6678],{"type":21,"tag":322,"props":6679,"children":6680},{"style":329},[6681],{"type":26,"value":6682},"    c.measure_all()\n",{"type":21,"tag":322,"props":6684,"children":6685},{"class":324,"line":1034},[6686,6691,6695],{"type":21,"tag":322,"props":6687,"children":6688},{"style":329},[6689],{"type":26,"value":6690},"    counts ",{"type":21,"tag":322,"props":6692,"children":6693},{"style":3638},[6694],{"type":26,"value":3678},{"type":21,"tag":322,"props":6696,"children":6697},{"style":329},[6698],{"type":26,"value":6699}," backend.run(\n",{"type":21,"tag":322,"props":6701,"children":6702},{"class":324,"line":3954},[6703,6708,6713,6717,6721,6725,6729,6733],{"type":21,"tag":322,"props":6704,"children":6705},{"style":329},[6706],{"type":26,"value":6707},"        transpile(c, backend, ",{"type":21,"tag":322,"props":6709,"children":6710},{"style":3703},[6711],{"type":26,"value":6712},"optimization_level",{"type":21,"tag":322,"props":6714,"children":6715},{"style":3638},[6716],{"type":26,"value":3678},{"type":21,"tag":322,"props":6718,"children":6719},{"style":348},[6720],{"type":26,"value":5605},{"type":21,"tag":322,"props":6722,"children":6723},{"style":329},[6724],{"type":26,"value":5775},{"type":21,"tag":322,"props":6726,"children":6727},{"style":3703},[6728],{"type":26,"value":4560},{"type":21,"tag":322,"props":6730,"children":6731},{"style":3638},[6732],{"type":26,"value":3678},{"type":21,"tag":322,"props":6734,"children":6735},{"style":329},[6736],{"type":26,"value":6737},"shots\n",{"type":21,"tag":322,"props":6739,"children":6740},{"class":324,"line":4002},[6741],{"type":21,"tag":322,"props":6742,"children":6743},{"style":329},[6744],{"type":26,"value":6745},"    ).result().get_counts()\n",{"type":21,"tag":322,"props":6747,"children":6748},{"class":324,"line":4044},[6749,6753,6758,6763,6767,6771,6775,6779,6784,6789,6793,6798,6803,6808,6812,6817,6822,6826,6831],{"type":21,"tag":322,"props":6750,"children":6751},{"style":3638},[6752],{"type":26,"value":5023},{"type":21,"tag":322,"props":6754,"children":6755},{"style":348},[6756],{"type":26,"value":6757}," sum",{"type":21,"tag":322,"props":6759,"children":6760},{"style":329},[6761],{"type":26,"value":6762},"((",{"type":21,"tag":322,"props":6764,"children":6765},{"style":3638},[6766],{"type":26,"value":5285},{"type":21,"tag":322,"props":6768,"children":6769},{"style":348},[6770],{"type":26,"value":3994},{"type":21,"tag":322,"props":6772,"children":6773},{"style":329},[6774],{"type":26,"value":5784},{"type":21,"tag":322,"props":6776,"children":6777},{"style":3638},[6778],{"type":26,"value":5150},{"type":21,"tag":322,"props":6780,"children":6781},{"style":329},[6782],{"type":26,"value":6783}," (k.replace(",{"type":21,"tag":322,"props":6785,"children":6786},{"style":359},[6787],{"type":26,"value":6788},"\" \"",{"type":21,"tag":322,"props":6790,"children":6791},{"style":329},[6792],{"type":26,"value":367},{"type":21,"tag":322,"props":6794,"children":6795},{"style":359},[6796],{"type":26,"value":6797},"\"\"",{"type":21,"tag":322,"props":6799,"children":6800},{"style":329},[6801],{"type":26,"value":6802},").count(",{"type":21,"tag":322,"props":6804,"children":6805},{"style":359},[6806],{"type":26,"value":6807},"\"1\"",{"type":21,"tag":322,"props":6809,"children":6810},{"style":329},[6811],{"type":26,"value":5784},{"type":21,"tag":322,"props":6813,"children":6814},{"style":3638},[6815],{"type":26,"value":6816},"%",{"type":21,"tag":322,"props":6818,"children":6819},{"style":348},[6820],{"type":26,"value":6821}," 2",{"type":21,"tag":322,"props":6823,"children":6824},{"style":329},[6825],{"type":26,"value":5784},{"type":21,"tag":322,"props":6827,"children":6828},{"style":3638},[6829],{"type":26,"value":6830},"*",{"type":21,"tag":322,"props":6832,"children":6833},{"style":329},[6834],{"type":26,"value":6835}," v\n",{"type":21,"tag":322,"props":6837,"children":6838},{"class":324,"line":5068},[6839,6844,6849,6853,6858,6862],{"type":21,"tag":322,"props":6840,"children":6841},{"style":3638},[6842],{"type":26,"value":6843},"               for",{"type":21,"tag":322,"props":6845,"children":6846},{"style":329},[6847],{"type":26,"value":6848}," k, v ",{"type":21,"tag":322,"props":6850,"children":6851},{"style":3638},[6852],{"type":26,"value":3941},{"type":21,"tag":322,"props":6854,"children":6855},{"style":329},[6856],{"type":26,"value":6857}," counts.items()) ",{"type":21,"tag":322,"props":6859,"children":6860},{"style":3638},[6861],{"type":26,"value":174},{"type":21,"tag":322,"props":6863,"children":6864},{"style":329},[6865],{"type":26,"value":5492},{"type":21,"tag":322,"props":6867,"children":6868},{"class":324,"line":5087},[6869],{"type":21,"tag":322,"props":6870,"children":6871},{"emptyLinePlaceholder":3662},[6872],{"type":26,"value":3665},{"type":21,"tag":322,"props":6874,"children":6875},{"class":324,"line":5095},[6876],{"type":21,"tag":322,"props":6877,"children":6878},{"style":3888},[6879],{"type":26,"value":6880},"# 4-qubit GHZ padded with cancelling CX pairs, ideal \u003CZZZZ> = +1.0\n",{"type":21,"tag":322,"props":6882,"children":6883},{"class":324,"line":5126},[6884,6889,6893,6897,6901],{"type":21,"tag":322,"props":6885,"children":6886},{"style":329},[6887],{"type":26,"value":6888},"base ",{"type":21,"tag":322,"props":6890,"children":6891},{"style":3638},[6892],{"type":26,"value":3678},{"type":21,"tag":322,"props":6894,"children":6895},{"style":329},[6896],{"type":26,"value":5575},{"type":21,"tag":322,"props":6898,"children":6899},{"style":348},[6900],{"type":26,"value":5969},{"type":21,"tag":322,"props":6902,"children":6903},{"style":329},[6904],{"type":26,"value":3738},{"type":21,"tag":322,"props":6906,"children":6907},{"class":324,"line":5171},[6908,6913,6917],{"type":21,"tag":322,"props":6909,"children":6910},{"style":329},[6911],{"type":26,"value":6912},"base.h(",{"type":21,"tag":322,"props":6914,"children":6915},{"style":348},[6916],{"type":26,"value":5605},{"type":21,"tag":322,"props":6918,"children":6919},{"style":329},[6920],{"type":26,"value":3738},{"type":21,"tag":322,"props":6922,"children":6923},{"class":324,"line":5210},[6924,6928,6932,6936,6940,6944,6949],{"type":21,"tag":322,"props":6925,"children":6926},{"style":3638},[6927],{"type":26,"value":3931},{"type":21,"tag":322,"props":6929,"children":6930},{"style":329},[6931],{"type":26,"value":3936},{"type":21,"tag":322,"props":6933,"children":6934},{"style":3638},[6935],{"type":26,"value":3941},{"type":21,"tag":322,"props":6937,"children":6938},{"style":348},[6939],{"type":26,"value":3946},{"type":21,"tag":322,"props":6941,"children":6942},{"style":329},[6943],{"type":26,"value":3758},{"type":21,"tag":322,"props":6945,"children":6946},{"style":348},[6947],{"type":26,"value":6948},"3",{"type":21,"tag":322,"props":6950,"children":6951},{"style":329},[6952],{"type":26,"value":4851},{"type":21,"tag":322,"props":6954,"children":6955},{"class":324,"line":5228},[6956,6961,6966,6970],{"type":21,"tag":322,"props":6957,"children":6958},{"style":329},[6959],{"type":26,"value":6960},"    base.cx(q, q ",{"type":21,"tag":322,"props":6962,"children":6963},{"style":3638},[6964],{"type":26,"value":6965},"+",{"type":21,"tag":322,"props":6967,"children":6968},{"style":348},[6969],{"type":26,"value":6536},{"type":21,"tag":322,"props":6971,"children":6972},{"style":329},[6973],{"type":26,"value":3738},{"type":21,"tag":322,"props":6975,"children":6976},{"class":324,"line":5302},[6977,6981,6985,6989,6993,6997,7001],{"type":21,"tag":322,"props":6978,"children":6979},{"style":3638},[6980],{"type":26,"value":3931},{"type":21,"tag":322,"props":6982,"children":6983},{"style":329},[6984],{"type":26,"value":6578},{"type":21,"tag":322,"props":6986,"children":6987},{"style":3638},[6988],{"type":26,"value":3941},{"type":21,"tag":322,"props":6990,"children":6991},{"style":348},[6992],{"type":26,"value":3946},{"type":21,"tag":322,"props":6994,"children":6995},{"style":329},[6996],{"type":26,"value":3758},{"type":21,"tag":322,"props":6998,"children":6999},{"style":348},[7000],{"type":26,"value":5969},{"type":21,"tag":322,"props":7002,"children":7003},{"style":329},[7004],{"type":26,"value":4851},{"type":21,"tag":322,"props":7006,"children":7007},{"class":324,"line":5330},[7008,7012,7016,7020,7024,7028,7032],{"type":21,"tag":322,"props":7009,"children":7010},{"style":3638},[7011],{"type":26,"value":5177},{"type":21,"tag":322,"props":7013,"children":7014},{"style":329},[7015],{"type":26,"value":3936},{"type":21,"tag":322,"props":7017,"children":7018},{"style":3638},[7019],{"type":26,"value":3941},{"type":21,"tag":322,"props":7021,"children":7022},{"style":348},[7023],{"type":26,"value":3946},{"type":21,"tag":322,"props":7025,"children":7026},{"style":329},[7027],{"type":26,"value":3758},{"type":21,"tag":322,"props":7029,"children":7030},{"style":348},[7031],{"type":26,"value":6948},{"type":21,"tag":322,"props":7033,"children":7034},{"style":329},[7035],{"type":26,"value":4851},{"type":21,"tag":322,"props":7037,"children":7038},{"class":324,"line":5359},[7039,7044,7048,7052],{"type":21,"tag":322,"props":7040,"children":7041},{"style":329},[7042],{"type":26,"value":7043},"        base.cx(q, q ",{"type":21,"tag":322,"props":7045,"children":7046},{"style":3638},[7047],{"type":26,"value":6965},{"type":21,"tag":322,"props":7049,"children":7050},{"style":348},[7051],{"type":26,"value":6536},{"type":21,"tag":322,"props":7053,"children":7054},{"style":329},[7055],{"type":26,"value":3738},{"type":21,"tag":322,"props":7057,"children":7058},{"class":324,"line":5368},[7059,7063,7067,7071],{"type":21,"tag":322,"props":7060,"children":7061},{"style":329},[7062],{"type":26,"value":7043},{"type":21,"tag":322,"props":7064,"children":7065},{"style":3638},[7066],{"type":26,"value":6965},{"type":21,"tag":322,"props":7068,"children":7069},{"style":348},[7070],{"type":26,"value":6536},{"type":21,"tag":322,"props":7072,"children":7073},{"style":329},[7074],{"type":26,"value":3738},{"type":21,"tag":322,"props":7076,"children":7077},{"class":324,"line":5396},[7078],{"type":21,"tag":322,"props":7079,"children":7080},{"emptyLinePlaceholder":3662},[7081],{"type":26,"value":3665},{"type":21,"tag":322,"props":7083,"children":7084},{"class":324,"line":5427},[7085,7090,7094,7099,7103,7107,7111,7115,7120],{"type":21,"tag":322,"props":7086,"children":7087},{"style":329},[7088],{"type":26,"value":7089},"scales ",{"type":21,"tag":322,"props":7091,"children":7092},{"style":3638},[7093],{"type":26,"value":3678},{"type":21,"tag":322,"props":7095,"children":7096},{"style":329},[7097],{"type":26,"value":7098}," [",{"type":21,"tag":322,"props":7100,"children":7101},{"style":348},[7102],{"type":26,"value":3994},{"type":21,"tag":322,"props":7104,"children":7105},{"style":329},[7106],{"type":26,"value":367},{"type":21,"tag":322,"props":7108,"children":7109},{"style":348},[7110],{"type":26,"value":6948},{"type":21,"tag":322,"props":7112,"children":7113},{"style":329},[7114],{"type":26,"value":367},{"type":21,"tag":322,"props":7116,"children":7117},{"style":348},[7118],{"type":26,"value":7119},"5",{"type":21,"tag":322,"props":7121,"children":7122},{"style":329},[7123],{"type":26,"value":7124},"]\n",{"type":21,"tag":322,"props":7126,"children":7127},{"class":324,"line":5449},[7128,7133,7137,7142,7146,7151,7155],{"type":21,"tag":322,"props":7129,"children":7130},{"style":329},[7131],{"type":26,"value":7132},"vals ",{"type":21,"tag":322,"props":7134,"children":7135},{"style":3638},[7136],{"type":26,"value":3678},{"type":21,"tag":322,"props":7138,"children":7139},{"style":329},[7140],{"type":26,"value":7141}," [expval_parity(fold_global(base, s), backend) ",{"type":21,"tag":322,"props":7143,"children":7144},{"style":3638},[7145],{"type":26,"value":3931},{"type":21,"tag":322,"props":7147,"children":7148},{"style":329},[7149],{"type":26,"value":7150}," s ",{"type":21,"tag":322,"props":7152,"children":7153},{"style":3638},[7154],{"type":26,"value":3941},{"type":21,"tag":322,"props":7156,"children":7157},{"style":329},[7158],{"type":26,"value":7159}," scales]\n",{"type":21,"tag":322,"props":7161,"children":7162},{"class":324,"line":5495},[7163,7168,7172,7177,7181,7185,7189,7194],{"type":21,"tag":322,"props":7164,"children":7165},{"style":329},[7166],{"type":26,"value":7167},"zne ",{"type":21,"tag":322,"props":7169,"children":7170},{"style":3638},[7171],{"type":26,"value":3678},{"type":21,"tag":322,"props":7173,"children":7174},{"style":329},[7175],{"type":26,"value":7176}," np.polyval(np.polyfit(scales, vals, ",{"type":21,"tag":322,"props":7178,"children":7179},{"style":348},[7180],{"type":26,"value":423},{"type":21,"tag":322,"props":7182,"children":7183},{"style":329},[7184],{"type":26,"value":5775},{"type":21,"tag":322,"props":7186,"children":7187},{"style":348},[7188],{"type":26,"value":5605},{"type":21,"tag":322,"props":7190,"children":7191},{"style":329},[7192],{"type":26,"value":7193},")   ",{"type":21,"tag":322,"props":7195,"children":7196},{"style":3888},[7197],{"type":26,"value":7198},"# Richardson, order 2\n",{"type":21,"tag":322,"props":7200,"children":7201},{"class":324,"line":5508},[7202,7206,7211,7216],{"type":21,"tag":322,"props":7203,"children":7204},{"style":348},[7205],{"type":26,"value":3753},{"type":21,"tag":322,"props":7207,"children":7208},{"style":329},[7209],{"type":26,"value":7210},"(vals, ",{"type":21,"tag":322,"props":7212,"children":7213},{"style":359},[7214],{"type":26,"value":7215},"\"->\"",{"type":21,"tag":322,"props":7217,"children":7218},{"style":329},[7219],{"type":26,"value":7220},", zne)\n",{"type":21,"tag":22,"props":7222,"children":7223},{},[7224,7229],{"type":21,"tag":71,"props":7225,"children":7226},{},[7227],{"type":26,"value":7228},"Real measured output",{"type":26,"value":7230},", same versions, readout error switched off to isolate gate noise. With a 0.4% two-qubit error rate:",{"type":21,"tag":312,"props":7232,"children":7234},{"code":7233},"scale 1: 0.8973   scale 3: 0.7286   scale 5: 0.5767\nZNE estimate: 0.988   (exact: 1.000)\n",[7235],{"type":21,"tag":79,"props":7236,"children":7237},{"__ignoreMap":7},[7238],{"type":26,"value":7233},{"type":21,"tag":22,"props":7240,"children":7241},{},[7242],{"type":26,"value":7243},"Raw error 0.103, mitigated error 0.012. ZNE removed 88% of the bias. This is what the technique looks like when it works.",{"type":21,"tag":59,"props":7245,"children":7247},{"id":7246},"when-zne-lies",[7248],{"type":26,"value":7249},"When ZNE Lies",{"type":21,"tag":22,"props":7251,"children":7252},{},[7253],{"type":26,"value":7254},"Now turn the two-qubit error rate up to 4% and rerun the identical code:",{"type":21,"tag":312,"props":7256,"children":7258},{"code":7257},"scale 1: 0.3311   scale 3: 0.0320   scale 5: 0.0008\nZNE estimate: 0.581   (exact: 1.000)\n",[7259],{"type":21,"tag":79,"props":7260,"children":7261},{"__ignoreMap":7},[7262],{"type":26,"value":7257},{"type":21,"tag":22,"props":7264,"children":7265},{},[7266,7268,7273],{"type":26,"value":7267},"Look at what a results table would show: raw 0.33, mitigated 0.58, ideal 1.00. Mitigation \"recovered\" 76% more signal! It looks like a triumph. It is garbage. The λ=5 point is 0.0008. The signal is ",{"type":21,"tag":51,"props":7269,"children":7270},{},[7271],{"type":26,"value":7272},"gone",{"type":26,"value":7274},". The fit is being driven by two nearly-zero numbers and one noisy one, and it happens to land somewhere between raw and truth because the curve shape forces it to.",{"type":21,"tag":22,"props":7276,"children":7277},{},[7278,7280,7286,7288,7293,7295,7300],{"type":26,"value":7279},"This is precisely the failure mode Köster and Mauerer characterize in ",{"type":21,"tag":29,"props":7281,"children":7283},{"href":7282},"\u002Fresearch\u002Fzne-artefactual-improvements-2026",[7284],{"type":26,"value":7285},"Artefactual Improvements in Zero-Noise Extrapolation",{"type":26,"value":7287}," (2026). Their finding: once noise amplification pushes past usable signal, ",{"type":21,"tag":71,"props":7289,"children":7290},{},[7291],{"type":26,"value":7292},"Richardson extrapolation degenerates into a fixed rescaling of a single noisy measurement",{"type":26,"value":7294},". The output is no longer a function of the noise amplification it supposedly corrects for. The amplified points contribute nothing but their own vanishing. On real hardware they measured runs that ",{"type":21,"tag":51,"props":7296,"children":7297},{},[7298],{"type":26,"value":7299},"overshot",{"type":26,"value":7301}," the ideal value by up to 21%.",{"type":21,"tag":22,"props":7303,"children":7304},{},[7305,7307,7312],{"type":26,"value":7306},"Their sharpest result is a negative control. They defined \"garbage folding\" (circuit modifications with no relationship to the identity-insertion ZNE requires, i.e. deliberately meaningless) and pushed those through the same pipeline. It produced ",{"type":21,"tag":51,"props":7308,"children":7309},{},[7310],{"type":26,"value":7311},"larger",{"type":26,"value":7313}," apparent improvements than legitimate folding. If nonsense scores better than the real method, the score is not measuring what you think it is.",{"type":21,"tag":22,"props":7315,"children":7316},{},[7317,7322],{"type":21,"tag":71,"props":7318,"children":7319},{},[7320],{"type":26,"value":7321},"The operational lesson: improvement magnitude never validates a mitigation result.",{"type":26,"value":7323}," A number moving toward the expected answer is not evidence the method worked. It might instead be evidence the method collapsed into a rescaling that happens to move numbers in that direction.",{"type":21,"tag":59,"props":7325,"children":7327},{"id":7326},"the-sanity-checks-that-catch-this",[7328],{"type":26,"value":7329},"The Sanity Checks That Catch This",{"type":21,"tag":22,"props":7331,"children":7332},{},[7333],{"type":26,"value":7334},"Three of them, none expensive:",{"type":21,"tag":22,"props":7336,"children":7337},{},[7338,7343,7345,7351],{"type":21,"tag":71,"props":7339,"children":7340},{},[7341],{"type":26,"value":7342},"1. Inspect the raw scale-factor points, always.",{"type":26,"value":7344}," If your highest-λ value has collapsed toward the fully-depolarized value (0 for a parity observable, 1\u002F2^n for a probability), discard it. Fitting through dead points is fitting noise. In the failing run above, one glance at ",{"type":21,"tag":79,"props":7346,"children":7348},{"className":7347},[],[7349],{"type":26,"value":7350},"0.0008",{"type":26,"value":7352}," tells you the extrapolation is unsupported.",{"type":21,"tag":22,"props":7354,"children":7355},{},[7356,7361,7363,7368,7370,7376,7378,7383],{"type":21,"tag":71,"props":7357,"children":7358},{},[7359],{"type":26,"value":7360},"2. Run a negative control.",{"type":26,"value":7362}," Build a variant of your folding that ",{"type":21,"tag":51,"props":7364,"children":7365},{},[7366],{"type":26,"value":7367},"shouldn't",{"type":26,"value":7369}," work (random gates of comparable depth instead of ",{"type":21,"tag":79,"props":7371,"children":7373},{"className":7372},[],[7374],{"type":26,"value":7375},"U†U",{"type":26,"value":7377}," pairs) and push it through the identical pipeline. In my failing case, garbage folding returned 0.186 against a raw value of 0.329: it moved the answer ",{"type":21,"tag":51,"props":7379,"children":7380},{},[7381],{"type":26,"value":7382},"away",{"type":26,"value":7384}," from truth. That divergence is the flag. Whether your control produces spurious improvement or spurious degradation, any large response to meaningless input means your pipeline is fitting artifacts.",{"type":21,"tag":22,"props":7386,"children":7387},{},[7388,7399],{"type":21,"tag":71,"props":7389,"children":7390},{},[7391,7393,7397],{"type":26,"value":7392},"3. Verify against a ",{"type":21,"tag":29,"props":7394,"children":7395},{"href":1005},[7396],{"type":26,"value":3724},{"type":26,"value":7398}," at small scale.",{"type":26,"value":7400}," Run the same mitigation pipeline on a problem size you compute exactly. If ZNE doesn't recover the known answer at 6 qubits, nothing about 60 qubits will be trustworthy.",{"type":21,"tag":59,"props":7402,"children":7404},{"id":7403},"probabilistic-error-cancellation-briefly",[7405],{"type":26,"value":7406},"Probabilistic Error Cancellation (Briefly)",{"type":21,"tag":22,"props":7408,"children":7409},{},[7410,7412,7417],{"type":26,"value":7411},"PEC is the rigorous option. Characterize the noise channel precisely, express the ",{"type":21,"tag":51,"props":7413,"children":7414},{},[7415],{"type":26,"value":7416},"inverse",{"type":26,"value":7418}," channel as a quasi-probability distribution over implementable operations, then sample from it with signed weights. It's unbiased in principle. No extrapolation guesswork.",{"type":21,"tag":22,"props":7420,"children":7421},{},[7422,7424,7430],{"type":26,"value":7423},"The cost is brutal. Sampling overhead scales as γ², where γ grows exponentially in circuit depth × error rate. A modest circuit sometimes needs 100-10,000 times more shots than the unmitigated run, on top of an expensive noise-characterization phase (gate set tomography or cycle benchmarking). PEC is appropriate for small, high-value circuits where you need a trustworthy number and afford the ",{"type":21,"tag":29,"props":7425,"children":7427},{"href":7426},"\u002Fglossary\u002Fshot-budget",[7428],{"type":26,"value":7429},"shot budget",{"type":26,"value":7431},". For most work, it isn't.",{"type":21,"tag":59,"props":7433,"children":7435},{"id":7434},"what-to-do",[7436],{"type":26,"value":7437},"What To Do",{"type":21,"tag":22,"props":7439,"children":7440},{},[7441],{"type":26,"value":7442},"In order:",{"type":21,"tag":7444,"props":7445,"children":7446},"ol",{},[7447,7457,7467,7477,7487],{"type":21,"tag":3560,"props":7448,"children":7449},{},[7450,7455],{"type":21,"tag":71,"props":7451,"children":7452},{},[7453],{"type":26,"value":7454},"Readout mitigation",{"type":26,"value":7456},": always. Cheap, bounded, low-risk.",{"type":21,"tag":3560,"props":7458,"children":7459},{},[7460,7465],{"type":21,"tag":71,"props":7461,"children":7462},{},[7463],{"type":26,"value":7464},"Dynamical decoupling",{"type":26,"value":7466},": always, if your circuit has idle qubits. Free.",{"type":21,"tag":3560,"props":7468,"children":7469},{},[7470,7475],{"type":21,"tag":71,"props":7471,"children":7472},{},[7473],{"type":26,"value":7474},"Twirling",{"type":26,"value":7476},": if you plan to use ZNE or PEC, so the noise is stochastic.",{"type":21,"tag":3560,"props":7478,"children":7479},{},[7480,7485],{"type":21,"tag":71,"props":7481,"children":7482},{},[7483],{"type":26,"value":7484},"ZNE",{"type":26,"value":7486},": only with scale-factor inspection and a negative control.",{"type":21,"tag":3560,"props":7488,"children":7489},{},[7490,7495],{"type":21,"tag":71,"props":7491,"children":7492},{},[7493],{"type":26,"value":7494},"PEC",{"type":26,"value":7496},": only when the circuit is small and the answer matters more than the cost.",{"type":21,"tag":22,"props":7498,"children":7499},{},[7500,7505,7507,7512,7514,7520],{"type":21,"tag":71,"props":7501,"children":7502},{},[7503],{"type":26,"value":7504},"Skip mitigation entirely when:",{"type":26,"value":7506}," your raw signal is already below ~0.2 of its ideal magnitude (nothing to extrapolate from), your circuit is deep enough that the λ=3 run is fully depolarized, or you're doing variational optimization where the optimizer only needs the ",{"type":21,"tag":51,"props":7508,"children":7509},{},[7510],{"type":26,"value":7511},"gradient direction",{"type":26,"value":7513},": bias that's roughly constant across parameter space often cancels out, and you're better off spending those shots on more iterations. Techniques for that are in ",{"type":21,"tag":29,"props":7515,"children":7517},{"href":7516},"\u002Fblog\u002Freduce-shot-count",[7518],{"type":26,"value":7519},"cutting your shot count",{"type":26,"value":534},{"type":21,"tag":22,"props":7522,"children":7523},{},[7524,7526,7532],{"type":26,"value":7525},"Mitigation buys you a few years of useful computation on machines that shouldn't be useful yet. It does not buy you fault tolerance, and it will happily hand you a confident wrong answer if you skip the controls. Get comfortable with the ",{"type":21,"tag":29,"props":7527,"children":7529},{"href":7528},"\u002Fsdks\u002Fqiskit",[7530],{"type":26,"value":7531},"Qiskit tooling",{"type":26,"value":7533},", always sanity-check against exact simulation, and treat any unvalidated improvement as a hypothesis rather than a result.",{"type":21,"tag":22,"props":7535,"children":7536},{},[7537,7542,7543,7548,7550,7555,7556],{"type":21,"tag":71,"props":7538,"children":7539},{},[7540],{"type":26,"value":7541},"Related:",{"type":26,"value":1901},{"type":21,"tag":29,"props":7544,"children":7545},{"href":854},[7546],{"type":26,"value":7547},"Error correction explained",{"type":26,"value":7549}," · ",{"type":21,"tag":29,"props":7551,"children":7552},{"href":7282},[7553],{"type":26,"value":7554},"ZNE artefacts paper",{"type":26,"value":7549},{"type":21,"tag":29,"props":7557,"children":7559},{"href":7558},"\u002Fglossary\u002Fquantum-utility",[7560],{"type":26,"value":7561},"Quantum utility",{"type":21,"tag":703,"props":7563,"children":7564},{},[7565],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":7567},[7568,7569,7570,7571,7572,7573,7574,7575,7576],{"id":4441,"depth":335,"text":4444},{"id":4638,"depth":335,"text":4641},{"id":6160,"depth":335,"text":6163},{"id":6400,"depth":335,"text":6403},{"id":6435,"depth":335,"text":6438},{"id":7246,"depth":335,"text":7249},{"id":7326,"depth":335,"text":7329},{"id":7403,"depth":335,"text":7406},{"id":7434,"depth":335,"text":7437},"content:blog:error-mitigation-practical-guide.md","blog\u002Ferror-mitigation-practical-guide.md","blog\u002Ferror-mitigation-practical-guide",{"_path":7581,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":7582,"description":7583,"date":3500,"author":11,"tags":7584,"readingTime":3884,"body":7587,"_type":715,"_id":9319,"_source":717,"_file":9320,"_stem":9321,"_extension":720},"\u002Fblog\u002Fgrover-algorithm-tutorial","Grover's Algorithm in Qiskit: A Practical Tutorial","Build Grover's search algorithm from scratch in Qiskit: oracle, diffuser, and the optimal iteration count, with complete runnable code and honest limits on the quadratic speedup.",[7585,1962,7586],"Algorithms","Beginners",{"type":18,"children":7588,"toc":9307},[7589,7600,7605,7611,7616,7621,7633,7639,7644,7674,7679,7685,7698,7850,7863,7869,7874,7895,8216,8244,8250,8255,8477,8490,8496,8501,8509,8519,8524,8942,8947,8955,8967,8973,8985,8990,8998,9010,9022,9028,9033,9051,9061,9085,9097,9103,9115,9183,9202,9221,9227,9303],{"type":21,"tag":22,"props":7590,"children":7591},{},[7592,7598],{"type":21,"tag":29,"props":7593,"children":7595},{"href":7594},"\u002Fglossary\u002Fgrover-algorithm",[7596],{"type":26,"value":7597},"Grover's algorithm",{"type":26,"value":7599}," is the second-most-famous quantum algorithm after Shor's, and it is by far the easier one to build. You implement a working version in about forty lines of Qiskit and watch the marked state emerge from the noise floor.",{"type":21,"tag":22,"props":7601,"children":7602},{},[7603],{"type":26,"value":7604},"It's also the algorithm most commonly misunderstood. This guide builds it step by step, shows the mistake that almost every beginner makes, and is honest about what the speedup does and does not buy you.",{"type":21,"tag":59,"props":7606,"children":7608},{"id":7607},"the-problem-unstructured-search",[7609],{"type":26,"value":7610},"The Problem: Unstructured Search",{"type":21,"tag":22,"props":7612,"children":7613},{},[7614],{"type":26,"value":7615},"You have N possibilities. Exactly one of them (or a few) satisfies some condition. You have no structure to exploit: no sorting, no index, no gradient, nothing that lets you rule out half the space with one look. Your only tool is a checker function: hand it a candidate, it tells you yes or no.",{"type":21,"tag":22,"props":7617,"children":7618},{},[7619],{"type":26,"value":7620},"Classically, you have no choice but to try candidates one at a time. On average that's N\u002F2 checks, and N in the worst case. Grover's algorithm finds the answer in roughly (π\u002F4)√N checks. For N = 1,000,000 that's about 785 instead of 500,000.",{"type":21,"tag":22,"props":7622,"children":7623},{},[7624,7626,7631],{"type":26,"value":7625},"The word ",{"type":21,"tag":71,"props":7627,"children":7628},{},[7629],{"type":26,"value":7630},"unstructured",{"type":26,"value":7632}," is doing enormous work here. It does not mean \"hard\". It means \"the only thing you do is test candidates\". If your problem has any exploitable structure (sorted data, a hash index, a metric that lets you prune), classical algorithms will use it and will crush Grover. The classic \"search a database of N records\" framing is genuinely misleading, and we'll come back to why.",{"type":21,"tag":59,"props":7634,"children":7636},{"id":7635},"the-core-idea-amplitude-amplification",[7637],{"type":26,"value":7638},"The Core Idea: Amplitude Amplification",{"type":21,"tag":22,"props":7640,"children":7641},{},[7642],{"type":26,"value":7643},"Start with every state equally likely. Then repeat two operations:",{"type":21,"tag":7444,"props":7645,"children":7646},{},[7647,7664],{"type":21,"tag":3560,"props":7648,"children":7649},{},[7650,7655,7657,7662],{"type":21,"tag":71,"props":7651,"children":7652},{},[7653],{"type":26,"value":7654},"The oracle",{"type":26,"value":7656}," flips the ",{"type":21,"tag":51,"props":7658,"children":7659},{},[7660],{"type":26,"value":7661},"sign",{"type":26,"value":7663}," of the amplitude of the marked state, leaving everything else alone. Probabilities don't change (all amplitudes are still ±1\u002F√N) but the marked one now points the other way.",{"type":21,"tag":3560,"props":7665,"children":7666},{},[7667,7672],{"type":21,"tag":71,"props":7668,"children":7669},{},[7670],{"type":26,"value":7671},"The diffusion operator",{"type":26,"value":7673}," reflects every amplitude about their mean. Since the marked amplitude is negative it sits far below the mean, so reflecting it sends it well above. Every other amplitude drops slightly.",{"type":21,"tag":22,"props":7675,"children":7676},{},[7677],{"type":26,"value":7678},"Each round is a rotation of the state vector toward the target by a fixed angle. That's the whole algorithm. The rotation angle is fixed, which is exactly why overshooting is possible. More on that shortly.",{"type":21,"tag":59,"props":7680,"children":7682},{"id":7681},"step-1-uniform-superposition",[7683],{"type":26,"value":7684},"Step 1: Uniform Superposition",{"type":21,"tag":22,"props":7686,"children":7687},{},[7688,7690,7696],{"type":26,"value":7689},"Apply a ",{"type":21,"tag":29,"props":7691,"children":7693},{"href":7692},"\u002Fglossary\u002Fhadamard-gate",[7694],{"type":26,"value":7695},"Hadamard gate",{"type":26,"value":7697}," to every qubit. Three qubits give eight basis states each with amplitude 1\u002F√8.",{"type":21,"tag":312,"props":7699,"children":7701},{"code":7700,"language":3628,"meta":7,"className":3626,"style":7},"from qiskit import QuantumCircuit, transpile\nfrom qiskit_aer import AerSimulator\nimport math\n\nn = 3                # 3 qubits -> N = 8 states\nmarked = \"101\"       # the state we want to find\n\nqc = QuantumCircuit(n, n)\nqc.h(range(n))       # uniform superposition\n",[7702],{"type":21,"tag":79,"props":7703,"children":7704},{"__ignoreMap":7},[7705,7724,7743,7755,7762,7783,7805,7812,7828],{"type":21,"tag":322,"props":7706,"children":7707},{"class":324,"line":325},[7708,7712,7716,7720],{"type":21,"tag":322,"props":7709,"children":7710},{"style":3638},[7711],{"type":26,"value":3641},{"type":21,"tag":322,"props":7713,"children":7714},{"style":329},[7715],{"type":26,"value":4732},{"type":21,"tag":322,"props":7717,"children":7718},{"style":3638},[7719],{"type":26,"value":3651},{"type":21,"tag":322,"props":7721,"children":7722},{"style":329},[7723],{"type":26,"value":4741},{"type":21,"tag":322,"props":7725,"children":7726},{"class":324,"line":335},[7727,7731,7735,7739],{"type":21,"tag":322,"props":7728,"children":7729},{"style":3638},[7730],{"type":26,"value":3641},{"type":21,"tag":322,"props":7732,"children":7733},{"style":329},[7734],{"type":26,"value":4753},{"type":21,"tag":322,"props":7736,"children":7737},{"style":3638},[7738],{"type":26,"value":3651},{"type":21,"tag":322,"props":7740,"children":7741},{"style":329},[7742],{"type":26,"value":4762},{"type":21,"tag":322,"props":7744,"children":7745},{"class":324,"line":344},[7746,7750],{"type":21,"tag":322,"props":7747,"children":7748},{"style":3638},[7749],{"type":26,"value":3651},{"type":21,"tag":322,"props":7751,"children":7752},{"style":329},[7753],{"type":26,"value":7754}," math\n",{"type":21,"tag":322,"props":7756,"children":7757},{"class":324,"line":407},[7758],{"type":21,"tag":322,"props":7759,"children":7760},{"emptyLinePlaceholder":3662},[7761],{"type":26,"value":3665},{"type":21,"tag":322,"props":7763,"children":7764},{"class":324,"line":431},[7765,7769,7773,7778],{"type":21,"tag":322,"props":7766,"children":7767},{"style":329},[7768],{"type":26,"value":5522},{"type":21,"tag":322,"props":7770,"children":7771},{"style":3638},[7772],{"type":26,"value":3678},{"type":21,"tag":322,"props":7774,"children":7775},{"style":348},[7776],{"type":26,"value":7777}," 3",{"type":21,"tag":322,"props":7779,"children":7780},{"style":3888},[7781],{"type":26,"value":7782},"                # 3 qubits -> N = 8 states\n",{"type":21,"tag":322,"props":7784,"children":7785},{"class":324,"line":453},[7786,7791,7795,7800],{"type":21,"tag":322,"props":7787,"children":7788},{"style":329},[7789],{"type":26,"value":7790},"marked ",{"type":21,"tag":322,"props":7792,"children":7793},{"style":3638},[7794],{"type":26,"value":3678},{"type":21,"tag":322,"props":7796,"children":7797},{"style":359},[7798],{"type":26,"value":7799}," \"101\"",{"type":21,"tag":322,"props":7801,"children":7802},{"style":3888},[7803],{"type":26,"value":7804},"       # the state we want to find\n",{"type":21,"tag":322,"props":7806,"children":7807},{"class":324,"line":474},[7808],{"type":21,"tag":322,"props":7809,"children":7810},{"emptyLinePlaceholder":3662},[7811],{"type":26,"value":3665},{"type":21,"tag":322,"props":7813,"children":7814},{"class":324,"line":496},[7815,7820,7824],{"type":21,"tag":322,"props":7816,"children":7817},{"style":329},[7818],{"type":26,"value":7819},"qc ",{"type":21,"tag":322,"props":7821,"children":7822},{"style":3638},[7823],{"type":26,"value":3678},{"type":21,"tag":322,"props":7825,"children":7826},{"style":329},[7827],{"type":26,"value":5225},{"type":21,"tag":322,"props":7829,"children":7830},{"class":324,"line":16},[7831,7836,7840,7845],{"type":21,"tag":322,"props":7832,"children":7833},{"style":329},[7834],{"type":26,"value":7835},"qc.h(",{"type":21,"tag":322,"props":7837,"children":7838},{"style":348},[7839],{"type":26,"value":5379},{"type":21,"tag":322,"props":7841,"children":7842},{"style":329},[7843],{"type":26,"value":7844},"(n))       ",{"type":21,"tag":322,"props":7846,"children":7847},{"style":3888},[7848],{"type":26,"value":7849},"# uniform superposition\n",{"type":21,"tag":22,"props":7851,"children":7852},{},[7853,7855,7861],{"type":26,"value":7854},"If you ",{"type":21,"tag":29,"props":7856,"children":7858},{"href":7857},"\u002Fglossary\u002Fmeasurement",[7859],{"type":26,"value":7860},"measure",{"type":26,"value":7862}," here you get a uniform random bitstring: 12.5% for each of the eight outcomes. That's the baseline to beat.",{"type":21,"tag":59,"props":7864,"children":7866},{"id":7865},"step-2-the-oracle",[7867],{"type":26,"value":7868},"Step 2: The Oracle",{"type":21,"tag":22,"props":7870,"children":7871},{},[7872],{"type":26,"value":7873},"The oracle must apply a phase of −1 to |101⟩ and +1 to everything else. The trick is a multi-controlled Z gate: it flips the sign of |111⟩ only. To mark a different pattern, sandwich it with X gates on the qubits that should be 0.",{"type":21,"tag":22,"props":7875,"children":7876},{},[7877,7879,7885,7887,7893],{"type":26,"value":7878},"Qiskit has no ",{"type":21,"tag":79,"props":7880,"children":7882},{"className":7881},[],[7883],{"type":26,"value":7884},"mcz",{"type":26,"value":7886},", so we build it the standard way: an H on the target, a multi-controlled X (",{"type":21,"tag":29,"props":7888,"children":7890},{"href":7889},"\u002Fglossary\u002Ftoffoli-gate",[7891],{"type":26,"value":7892},"Toffoli",{"type":26,"value":7894}," generalized), an H back:",{"type":21,"tag":312,"props":7896,"children":7898},{"code":7897,"language":3628,"meta":7,"className":3626,"style":7},"def oracle(n, marked):\n    \"\"\"Phase-flip the marked basis state.\"\"\"\n    qc = QuantumCircuit(n, name=\"Oracle\")\n    rev = marked[::-1]              # Qiskit is little-endian\n    for i, bit in enumerate(rev):\n        if bit == \"0\":\n            qc.x(i)\n    qc.h(n - 1)                     # multi-controlled Z\n    qc.mcx(list(range(n - 1)), n - 1)\n    qc.h(n - 1)\n    for i, bit in enumerate(rev):\n        if bit == \"0\":\n            qc.x(i)\n    return qc\n",[7899],{"type":21,"tag":79,"props":7900,"children":7901},{"__ignoreMap":7},[7902,7919,7927,7962,7996,8021,8047,8055,8081,8132,8151,8174,8197,8204],{"type":21,"tag":322,"props":7903,"children":7904},{"class":324,"line":325},[7905,7909,7914],{"type":21,"tag":322,"props":7906,"children":7907},{"style":3638},[7908],{"type":26,"value":4798},{"type":21,"tag":322,"props":7910,"children":7911},{"style":4801},[7912],{"type":26,"value":7913}," oracle",{"type":21,"tag":322,"props":7915,"children":7916},{"style":329},[7917],{"type":26,"value":7918},"(n, marked):\n",{"type":21,"tag":322,"props":7920,"children":7921},{"class":324,"line":335},[7922],{"type":21,"tag":322,"props":7923,"children":7924},{"style":359},[7925],{"type":26,"value":7926},"    \"\"\"Phase-flip the marked basis state.\"\"\"\n",{"type":21,"tag":322,"props":7928,"children":7929},{"class":324,"line":344},[7930,7935,7939,7944,7949,7953,7958],{"type":21,"tag":322,"props":7931,"children":7932},{"style":329},[7933],{"type":26,"value":7934},"    qc ",{"type":21,"tag":322,"props":7936,"children":7937},{"style":3638},[7938],{"type":26,"value":3678},{"type":21,"tag":322,"props":7940,"children":7941},{"style":329},[7942],{"type":26,"value":7943}," QuantumCircuit(n, ",{"type":21,"tag":322,"props":7945,"children":7946},{"style":3703},[7947],{"type":26,"value":7948},"name",{"type":21,"tag":322,"props":7950,"children":7951},{"style":3638},[7952],{"type":26,"value":3678},{"type":21,"tag":322,"props":7954,"children":7955},{"style":359},[7956],{"type":26,"value":7957},"\"Oracle\"",{"type":21,"tag":322,"props":7959,"children":7960},{"style":329},[7961],{"type":26,"value":3738},{"type":21,"tag":322,"props":7963,"children":7964},{"class":324,"line":407},[7965,7970,7974,7979,7983,7987,7992],{"type":21,"tag":322,"props":7966,"children":7967},{"style":329},[7968],{"type":26,"value":7969},"    rev ",{"type":21,"tag":322,"props":7971,"children":7972},{"style":3638},[7973],{"type":26,"value":3678},{"type":21,"tag":322,"props":7975,"children":7976},{"style":329},[7977],{"type":26,"value":7978}," marked[::",{"type":21,"tag":322,"props":7980,"children":7981},{"style":3638},[7982],{"type":26,"value":5285},{"type":21,"tag":322,"props":7984,"children":7985},{"style":348},[7986],{"type":26,"value":3994},{"type":21,"tag":322,"props":7988,"children":7989},{"style":329},[7990],{"type":26,"value":7991},"]              ",{"type":21,"tag":322,"props":7993,"children":7994},{"style":3888},[7995],{"type":26,"value":5299},{"type":21,"tag":322,"props":7997,"children":7998},{"class":324,"line":431},[7999,8003,8008,8012,8016],{"type":21,"tag":322,"props":8000,"children":8001},{"style":3638},[8002],{"type":26,"value":5177},{"type":21,"tag":322,"props":8004,"children":8005},{"style":329},[8006],{"type":26,"value":8007}," i, bit ",{"type":21,"tag":322,"props":8009,"children":8010},{"style":3638},[8011],{"type":26,"value":3941},{"type":21,"tag":322,"props":8013,"children":8014},{"style":348},[8015],{"type":26,"value":5322},{"type":21,"tag":322,"props":8017,"children":8018},{"style":329},[8019],{"type":26,"value":8020},"(rev):\n",{"type":21,"tag":322,"props":8022,"children":8023},{"class":324,"line":453},[8024,8029,8034,8038,8043],{"type":21,"tag":322,"props":8025,"children":8026},{"style":3638},[8027],{"type":26,"value":8028},"        if",{"type":21,"tag":322,"props":8030,"children":8031},{"style":329},[8032],{"type":26,"value":8033}," bit ",{"type":21,"tag":322,"props":8035,"children":8036},{"style":3638},[8037],{"type":26,"value":5346},{"type":21,"tag":322,"props":8039,"children":8040},{"style":359},[8041],{"type":26,"value":8042}," \"0\"",{"type":21,"tag":322,"props":8044,"children":8045},{"style":329},[8046],{"type":26,"value":5356},{"type":21,"tag":322,"props":8048,"children":8049},{"class":324,"line":474},[8050],{"type":21,"tag":322,"props":8051,"children":8052},{"style":329},[8053],{"type":26,"value":8054},"            qc.x(i)\n",{"type":21,"tag":322,"props":8056,"children":8057},{"class":324,"line":496},[8058,8063,8067,8071,8076],{"type":21,"tag":322,"props":8059,"children":8060},{"style":329},[8061],{"type":26,"value":8062},"    qc.h(n ",{"type":21,"tag":322,"props":8064,"children":8065},{"style":3638},[8066],{"type":26,"value":5285},{"type":21,"tag":322,"props":8068,"children":8069},{"style":348},[8070],{"type":26,"value":6536},{"type":21,"tag":322,"props":8072,"children":8073},{"style":329},[8074],{"type":26,"value":8075},")                     ",{"type":21,"tag":322,"props":8077,"children":8078},{"style":3888},[8079],{"type":26,"value":8080},"# multi-controlled Z\n",{"type":21,"tag":322,"props":8082,"children":8083},{"class":324,"line":16},[8084,8089,8094,8098,8102,8107,8111,8115,8120,8124,8128],{"type":21,"tag":322,"props":8085,"children":8086},{"style":329},[8087],{"type":26,"value":8088},"    qc.mcx(",{"type":21,"tag":322,"props":8090,"children":8091},{"style":348},[8092],{"type":26,"value":8093},"list",{"type":21,"tag":322,"props":8095,"children":8096},{"style":329},[8097],{"type":26,"value":3758},{"type":21,"tag":322,"props":8099,"children":8100},{"style":348},[8101],{"type":26,"value":5379},{"type":21,"tag":322,"props":8103,"children":8104},{"style":329},[8105],{"type":26,"value":8106},"(n ",{"type":21,"tag":322,"props":8108,"children":8109},{"style":3638},[8110],{"type":26,"value":5285},{"type":21,"tag":322,"props":8112,"children":8113},{"style":348},[8114],{"type":26,"value":6536},{"type":21,"tag":322,"props":8116,"children":8117},{"style":329},[8118],{"type":26,"value":8119},")), n ",{"type":21,"tag":322,"props":8121,"children":8122},{"style":3638},[8123],{"type":26,"value":5285},{"type":21,"tag":322,"props":8125,"children":8126},{"style":348},[8127],{"type":26,"value":6536},{"type":21,"tag":322,"props":8129,"children":8130},{"style":329},[8131],{"type":26,"value":3738},{"type":21,"tag":322,"props":8133,"children":8134},{"class":324,"line":3884},[8135,8139,8143,8147],{"type":21,"tag":322,"props":8136,"children":8137},{"style":329},[8138],{"type":26,"value":8062},{"type":21,"tag":322,"props":8140,"children":8141},{"style":3638},[8142],{"type":26,"value":5285},{"type":21,"tag":322,"props":8144,"children":8145},{"style":348},[8146],{"type":26,"value":6536},{"type":21,"tag":322,"props":8148,"children":8149},{"style":329},[8150],{"type":26,"value":3738},{"type":21,"tag":322,"props":8152,"children":8153},{"class":324,"line":3894},[8154,8158,8162,8166,8170],{"type":21,"tag":322,"props":8155,"children":8156},{"style":3638},[8157],{"type":26,"value":5177},{"type":21,"tag":322,"props":8159,"children":8160},{"style":329},[8161],{"type":26,"value":8007},{"type":21,"tag":322,"props":8163,"children":8164},{"style":3638},[8165],{"type":26,"value":3941},{"type":21,"tag":322,"props":8167,"children":8168},{"style":348},[8169],{"type":26,"value":5322},{"type":21,"tag":322,"props":8171,"children":8172},{"style":329},[8173],{"type":26,"value":8020},{"type":21,"tag":322,"props":8175,"children":8176},{"class":324,"line":1034},[8177,8181,8185,8189,8193],{"type":21,"tag":322,"props":8178,"children":8179},{"style":3638},[8180],{"type":26,"value":8028},{"type":21,"tag":322,"props":8182,"children":8183},{"style":329},[8184],{"type":26,"value":8033},{"type":21,"tag":322,"props":8186,"children":8187},{"style":3638},[8188],{"type":26,"value":5346},{"type":21,"tag":322,"props":8190,"children":8191},{"style":359},[8192],{"type":26,"value":8042},{"type":21,"tag":322,"props":8194,"children":8195},{"style":329},[8196],{"type":26,"value":5356},{"type":21,"tag":322,"props":8198,"children":8199},{"class":324,"line":3954},[8200],{"type":21,"tag":322,"props":8201,"children":8202},{"style":329},[8203],{"type":26,"value":8054},{"type":21,"tag":322,"props":8205,"children":8206},{"class":324,"line":4002},[8207,8211],{"type":21,"tag":322,"props":8208,"children":8209},{"style":3638},[8210],{"type":26,"value":5023},{"type":21,"tag":322,"props":8212,"children":8213},{"style":329},[8214],{"type":26,"value":8215}," qc\n",{"type":21,"tag":22,"props":8217,"children":8218},{},[8219,8221,8226,8228,8234,8236,8242],{"type":26,"value":8220},"Watch the endianness. Qiskit orders bitstrings with qubit 0 on the ",{"type":21,"tag":71,"props":8222,"children":8223},{},[8224],{"type":26,"value":8225},"right",{"type":26,"value":8227},", so ",{"type":21,"tag":79,"props":8229,"children":8231},{"className":8230},[],[8232],{"type":26,"value":8233},"marked[::-1]",{"type":26,"value":8235}," maps the string to qubit indices. Getting this backwards is the second most common bug in Grover implementations and it fails silently. You'll amplify ",{"type":21,"tag":79,"props":8237,"children":8239},{"className":8238},[],[8240],{"type":26,"value":8241},"101",{"type":26,"value":8243}," reversed and wonder why.",{"type":21,"tag":59,"props":8245,"children":8247},{"id":8246},"step-3-the-diffusion-operator",[8248],{"type":26,"value":8249},"Step 3: The Diffusion Operator",{"type":21,"tag":22,"props":8251,"children":8252},{},[8253],{"type":26,"value":8254},"Inversion about the mean is H⊗n, then X⊗n, then a multi-controlled Z, then X⊗n, then H⊗n. Structurally it's a reflection about the uniform superposition state:",{"type":21,"tag":312,"props":8256,"children":8258},{"code":8257,"language":3628,"meta":7,"className":3626,"style":7},"def diffuser(n):\n    \"\"\"Inversion about the mean.\"\"\"\n    qc = QuantumCircuit(n, name=\"Diffuser\")\n    qc.h(range(n))\n    qc.x(range(n))\n    qc.h(n - 1)\n    qc.mcx(list(range(n - 1)), n - 1)\n    qc.h(n - 1)\n    qc.x(range(n))\n    qc.h(range(n))\n    return qc\n",[8259],{"type":21,"tag":79,"props":8260,"children":8261},{"__ignoreMap":7},[8262,8279,8287,8319,8335,8351,8370,8417,8436,8451,8466],{"type":21,"tag":322,"props":8263,"children":8264},{"class":324,"line":325},[8265,8269,8274],{"type":21,"tag":322,"props":8266,"children":8267},{"style":3638},[8268],{"type":26,"value":4798},{"type":21,"tag":322,"props":8270,"children":8271},{"style":4801},[8272],{"type":26,"value":8273}," diffuser",{"type":21,"tag":322,"props":8275,"children":8276},{"style":329},[8277],{"type":26,"value":8278},"(n):\n",{"type":21,"tag":322,"props":8280,"children":8281},{"class":324,"line":335},[8282],{"type":21,"tag":322,"props":8283,"children":8284},{"style":359},[8285],{"type":26,"value":8286},"    \"\"\"Inversion about the mean.\"\"\"\n",{"type":21,"tag":322,"props":8288,"children":8289},{"class":324,"line":344},[8290,8294,8298,8302,8306,8310,8315],{"type":21,"tag":322,"props":8291,"children":8292},{"style":329},[8293],{"type":26,"value":7934},{"type":21,"tag":322,"props":8295,"children":8296},{"style":3638},[8297],{"type":26,"value":3678},{"type":21,"tag":322,"props":8299,"children":8300},{"style":329},[8301],{"type":26,"value":7943},{"type":21,"tag":322,"props":8303,"children":8304},{"style":3703},[8305],{"type":26,"value":7948},{"type":21,"tag":322,"props":8307,"children":8308},{"style":3638},[8309],{"type":26,"value":3678},{"type":21,"tag":322,"props":8311,"children":8312},{"style":359},[8313],{"type":26,"value":8314},"\"Diffuser\"",{"type":21,"tag":322,"props":8316,"children":8317},{"style":329},[8318],{"type":26,"value":3738},{"type":21,"tag":322,"props":8320,"children":8321},{"class":324,"line":407},[8322,8327,8331],{"type":21,"tag":322,"props":8323,"children":8324},{"style":329},[8325],{"type":26,"value":8326},"    qc.h(",{"type":21,"tag":322,"props":8328,"children":8329},{"style":348},[8330],{"type":26,"value":5379},{"type":21,"tag":322,"props":8332,"children":8333},{"style":329},[8334],{"type":26,"value":5393},{"type":21,"tag":322,"props":8336,"children":8337},{"class":324,"line":431},[8338,8343,8347],{"type":21,"tag":322,"props":8339,"children":8340},{"style":329},[8341],{"type":26,"value":8342},"    qc.x(",{"type":21,"tag":322,"props":8344,"children":8345},{"style":348},[8346],{"type":26,"value":5379},{"type":21,"tag":322,"props":8348,"children":8349},{"style":329},[8350],{"type":26,"value":5393},{"type":21,"tag":322,"props":8352,"children":8353},{"class":324,"line":453},[8354,8358,8362,8366],{"type":21,"tag":322,"props":8355,"children":8356},{"style":329},[8357],{"type":26,"value":8062},{"type":21,"tag":322,"props":8359,"children":8360},{"style":3638},[8361],{"type":26,"value":5285},{"type":21,"tag":322,"props":8363,"children":8364},{"style":348},[8365],{"type":26,"value":6536},{"type":21,"tag":322,"props":8367,"children":8368},{"style":329},[8369],{"type":26,"value":3738},{"type":21,"tag":322,"props":8371,"children":8372},{"class":324,"line":474},[8373,8377,8381,8385,8389,8393,8397,8401,8405,8409,8413],{"type":21,"tag":322,"props":8374,"children":8375},{"style":329},[8376],{"type":26,"value":8088},{"type":21,"tag":322,"props":8378,"children":8379},{"style":348},[8380],{"type":26,"value":8093},{"type":21,"tag":322,"props":8382,"children":8383},{"style":329},[8384],{"type":26,"value":3758},{"type":21,"tag":322,"props":8386,"children":8387},{"style":348},[8388],{"type":26,"value":5379},{"type":21,"tag":322,"props":8390,"children":8391},{"style":329},[8392],{"type":26,"value":8106},{"type":21,"tag":322,"props":8394,"children":8395},{"style":3638},[8396],{"type":26,"value":5285},{"type":21,"tag":322,"props":8398,"children":8399},{"style":348},[8400],{"type":26,"value":6536},{"type":21,"tag":322,"props":8402,"children":8403},{"style":329},[8404],{"type":26,"value":8119},{"type":21,"tag":322,"props":8406,"children":8407},{"style":3638},[8408],{"type":26,"value":5285},{"type":21,"tag":322,"props":8410,"children":8411},{"style":348},[8412],{"type":26,"value":6536},{"type":21,"tag":322,"props":8414,"children":8415},{"style":329},[8416],{"type":26,"value":3738},{"type":21,"tag":322,"props":8418,"children":8419},{"class":324,"line":496},[8420,8424,8428,8432],{"type":21,"tag":322,"props":8421,"children":8422},{"style":329},[8423],{"type":26,"value":8062},{"type":21,"tag":322,"props":8425,"children":8426},{"style":3638},[8427],{"type":26,"value":5285},{"type":21,"tag":322,"props":8429,"children":8430},{"style":348},[8431],{"type":26,"value":6536},{"type":21,"tag":322,"props":8433,"children":8434},{"style":329},[8435],{"type":26,"value":3738},{"type":21,"tag":322,"props":8437,"children":8438},{"class":324,"line":16},[8439,8443,8447],{"type":21,"tag":322,"props":8440,"children":8441},{"style":329},[8442],{"type":26,"value":8342},{"type":21,"tag":322,"props":8444,"children":8445},{"style":348},[8446],{"type":26,"value":5379},{"type":21,"tag":322,"props":8448,"children":8449},{"style":329},[8450],{"type":26,"value":5393},{"type":21,"tag":322,"props":8452,"children":8453},{"class":324,"line":3884},[8454,8458,8462],{"type":21,"tag":322,"props":8455,"children":8456},{"style":329},[8457],{"type":26,"value":8326},{"type":21,"tag":322,"props":8459,"children":8460},{"style":348},[8461],{"type":26,"value":5379},{"type":21,"tag":322,"props":8463,"children":8464},{"style":329},[8465],{"type":26,"value":5393},{"type":21,"tag":322,"props":8467,"children":8468},{"class":324,"line":3894},[8469,8473],{"type":21,"tag":322,"props":8470,"children":8471},{"style":3638},[8472],{"type":26,"value":5023},{"type":21,"tag":322,"props":8474,"children":8475},{"style":329},[8476],{"type":26,"value":8215},{"type":21,"tag":22,"props":8478,"children":8479},{},[8480,8482,8488],{"type":26,"value":8481},"Note the diffuser knows nothing about the marked state. It is the same ",{"type":21,"tag":29,"props":8483,"children":8485},{"href":8484},"\u002Fglossary\u002Fquantum-circuit",[8486],{"type":26,"value":8487},"circuit",{"type":26,"value":8489}," for every search problem. Only the oracle changes.",{"type":21,"tag":59,"props":8491,"children":8493},{"id":8492},"step-4-iterate-the-right-number-of-times",[8494],{"type":26,"value":8495},"Step 4: Iterate the Right Number of Times",{"type":21,"tag":22,"props":8497,"children":8498},{},[8499],{"type":26,"value":8500},"This is where people go wrong. The optimal count is:",{"type":21,"tag":312,"props":8502,"children":8504},{"code":8503},"k = floor( (π\u002F4) · √(N\u002FM) )\n",[8505],{"type":21,"tag":79,"props":8506,"children":8507},{"__ignoreMap":7},[8508],{"type":26,"value":8503},{"type":21,"tag":22,"props":8510,"children":8511},{},[8512,8514,8518],{"type":26,"value":8513},"where N = 2ⁿ and M is the number of marked states. For N = 8, M = 1: k = floor(2.22) = ",{"type":21,"tag":71,"props":8515,"children":8516},{},[8517],{"type":26,"value":423},{"type":26,"value":534},{"type":21,"tag":22,"props":8520,"children":8521},{},[8522],{"type":26,"value":8523},"Here's the complete program:",{"type":21,"tag":312,"props":8525,"children":8527},{"code":8526,"language":3628,"meta":7,"className":3626,"style":7},"def grover(n, marked, iterations):\n    qc = QuantumCircuit(n, n)\n    qc.h(range(n))                       # uniform superposition\n    orc, dif = oracle(n, marked), diffuser(n)\n    for _ in range(iterations):\n        qc.compose(orc, inplace=True)    # phase-flip the target\n        qc.compose(dif, inplace=True)    # amplify it\n    qc.measure(range(n), range(n))\n    return qc\n\nN = 2 ** n\noptimal = math.floor(math.pi \u002F 4 * math.sqrt(N))\nprint(f\"N = {N}, optimal iterations = {optimal}\")\n\nsim = AerSimulator()\nqc = grover(n, marked, optimal)\ncounts = sim.run(transpile(qc, sim), shots=4096).result().get_counts()\nprint(sorted(counts.items()))\n",[8528],{"type":21,"tag":79,"props":8529,"children":8530},{"__ignoreMap":7},[8531,8548,8563,8583,8600,8624,8655,8684,8708,8719,8726,8752,8788,8847,8854,8871,8887,8921],{"type":21,"tag":322,"props":8532,"children":8533},{"class":324,"line":325},[8534,8538,8543],{"type":21,"tag":322,"props":8535,"children":8536},{"style":3638},[8537],{"type":26,"value":4798},{"type":21,"tag":322,"props":8539,"children":8540},{"style":4801},[8541],{"type":26,"value":8542}," grover",{"type":21,"tag":322,"props":8544,"children":8545},{"style":329},[8546],{"type":26,"value":8547},"(n, marked, iterations):\n",{"type":21,"tag":322,"props":8549,"children":8550},{"class":324,"line":335},[8551,8555,8559],{"type":21,"tag":322,"props":8552,"children":8553},{"style":329},[8554],{"type":26,"value":7934},{"type":21,"tag":322,"props":8556,"children":8557},{"style":3638},[8558],{"type":26,"value":3678},{"type":21,"tag":322,"props":8560,"children":8561},{"style":329},[8562],{"type":26,"value":5225},{"type":21,"tag":322,"props":8564,"children":8565},{"class":324,"line":344},[8566,8570,8574,8579],{"type":21,"tag":322,"props":8567,"children":8568},{"style":329},[8569],{"type":26,"value":8326},{"type":21,"tag":322,"props":8571,"children":8572},{"style":348},[8573],{"type":26,"value":5379},{"type":21,"tag":322,"props":8575,"children":8576},{"style":329},[8577],{"type":26,"value":8578},"(n))                       ",{"type":21,"tag":322,"props":8580,"children":8581},{"style":3888},[8582],{"type":26,"value":7849},{"type":21,"tag":322,"props":8584,"children":8585},{"class":324,"line":407},[8586,8591,8595],{"type":21,"tag":322,"props":8587,"children":8588},{"style":329},[8589],{"type":26,"value":8590},"    orc, dif ",{"type":21,"tag":322,"props":8592,"children":8593},{"style":3638},[8594],{"type":26,"value":3678},{"type":21,"tag":322,"props":8596,"children":8597},{"style":329},[8598],{"type":26,"value":8599}," oracle(n, marked), diffuser(n)\n",{"type":21,"tag":322,"props":8601,"children":8602},{"class":324,"line":431},[8603,8607,8611,8615,8619],{"type":21,"tag":322,"props":8604,"children":8605},{"style":3638},[8606],{"type":26,"value":5177},{"type":21,"tag":322,"props":8608,"children":8609},{"style":329},[8610],{"type":26,"value":6578},{"type":21,"tag":322,"props":8612,"children":8613},{"style":3638},[8614],{"type":26,"value":3941},{"type":21,"tag":322,"props":8616,"children":8617},{"style":348},[8618],{"type":26,"value":3946},{"type":21,"tag":322,"props":8620,"children":8621},{"style":329},[8622],{"type":26,"value":8623},"(iterations):\n",{"type":21,"tag":322,"props":8625,"children":8626},{"class":324,"line":453},[8627,8632,8637,8641,8645,8650],{"type":21,"tag":322,"props":8628,"children":8629},{"style":329},[8630],{"type":26,"value":8631},"        qc.compose(orc, ",{"type":21,"tag":322,"props":8633,"children":8634},{"style":3703},[8635],{"type":26,"value":8636},"inplace",{"type":21,"tag":322,"props":8638,"children":8639},{"style":3638},[8640],{"type":26,"value":3678},{"type":21,"tag":322,"props":8642,"children":8643},{"style":348},[8644],{"type":26,"value":3715},{"type":21,"tag":322,"props":8646,"children":8647},{"style":329},[8648],{"type":26,"value":8649},")    ",{"type":21,"tag":322,"props":8651,"children":8652},{"style":3888},[8653],{"type":26,"value":8654},"# phase-flip the target\n",{"type":21,"tag":322,"props":8656,"children":8657},{"class":324,"line":474},[8658,8663,8667,8671,8675,8679],{"type":21,"tag":322,"props":8659,"children":8660},{"style":329},[8661],{"type":26,"value":8662},"        qc.compose(dif, ",{"type":21,"tag":322,"props":8664,"children":8665},{"style":3703},[8666],{"type":26,"value":8636},{"type":21,"tag":322,"props":8668,"children":8669},{"style":3638},[8670],{"type":26,"value":3678},{"type":21,"tag":322,"props":8672,"children":8673},{"style":348},[8674],{"type":26,"value":3715},{"type":21,"tag":322,"props":8676,"children":8677},{"style":329},[8678],{"type":26,"value":8649},{"type":21,"tag":322,"props":8680,"children":8681},{"style":3888},[8682],{"type":26,"value":8683},"# amplify it\n",{"type":21,"tag":322,"props":8685,"children":8686},{"class":324,"line":496},[8687,8692,8696,8700,8704],{"type":21,"tag":322,"props":8688,"children":8689},{"style":329},[8690],{"type":26,"value":8691},"    qc.measure(",{"type":21,"tag":322,"props":8693,"children":8694},{"style":348},[8695],{"type":26,"value":5379},{"type":21,"tag":322,"props":8697,"children":8698},{"style":329},[8699],{"type":26,"value":5384},{"type":21,"tag":322,"props":8701,"children":8702},{"style":348},[8703],{"type":26,"value":5379},{"type":21,"tag":322,"props":8705,"children":8706},{"style":329},[8707],{"type":26,"value":5393},{"type":21,"tag":322,"props":8709,"children":8710},{"class":324,"line":16},[8711,8715],{"type":21,"tag":322,"props":8712,"children":8713},{"style":3638},[8714],{"type":26,"value":5023},{"type":21,"tag":322,"props":8716,"children":8717},{"style":329},[8718],{"type":26,"value":8215},{"type":21,"tag":322,"props":8720,"children":8721},{"class":324,"line":3884},[8722],{"type":21,"tag":322,"props":8723,"children":8724},{"emptyLinePlaceholder":3662},[8725],{"type":26,"value":3665},{"type":21,"tag":322,"props":8727,"children":8728},{"class":324,"line":3894},[8729,8734,8738,8742,8747],{"type":21,"tag":322,"props":8730,"children":8731},{"style":329},[8732],{"type":26,"value":8733},"N ",{"type":21,"tag":322,"props":8735,"children":8736},{"style":3638},[8737],{"type":26,"value":3678},{"type":21,"tag":322,"props":8739,"children":8740},{"style":348},[8741],{"type":26,"value":6821},{"type":21,"tag":322,"props":8743,"children":8744},{"style":3638},[8745],{"type":26,"value":8746}," **",{"type":21,"tag":322,"props":8748,"children":8749},{"style":329},[8750],{"type":26,"value":8751}," n\n",{"type":21,"tag":322,"props":8753,"children":8754},{"class":324,"line":1034},[8755,8760,8764,8769,8773,8778,8783],{"type":21,"tag":322,"props":8756,"children":8757},{"style":329},[8758],{"type":26,"value":8759},"optimal ",{"type":21,"tag":322,"props":8761,"children":8762},{"style":3638},[8763],{"type":26,"value":3678},{"type":21,"tag":322,"props":8765,"children":8766},{"style":329},[8767],{"type":26,"value":8768}," math.floor(math.pi ",{"type":21,"tag":322,"props":8770,"children":8771},{"style":3638},[8772],{"type":26,"value":174},{"type":21,"tag":322,"props":8774,"children":8775},{"style":348},[8776],{"type":26,"value":8777}," 4",{"type":21,"tag":322,"props":8779,"children":8780},{"style":3638},[8781],{"type":26,"value":8782}," *",{"type":21,"tag":322,"props":8784,"children":8785},{"style":329},[8786],{"type":26,"value":8787}," math.sqrt(N))\n",{"type":21,"tag":322,"props":8789,"children":8790},{"class":324,"line":3954},[8791,8795,8799,8803,8808,8812,8817,8821,8826,8830,8835,8839,8843],{"type":21,"tag":322,"props":8792,"children":8793},{"style":348},[8794],{"type":26,"value":3753},{"type":21,"tag":322,"props":8796,"children":8797},{"style":329},[8798],{"type":26,"value":3758},{"type":21,"tag":322,"props":8800,"children":8801},{"style":3638},[8802],{"type":26,"value":3763},{"type":21,"tag":322,"props":8804,"children":8805},{"style":359},[8806],{"type":26,"value":8807},"\"N = ",{"type":21,"tag":322,"props":8809,"children":8810},{"style":348},[8811],{"type":26,"value":3773},{"type":21,"tag":322,"props":8813,"children":8814},{"style":329},[8815],{"type":26,"value":8816},"N",{"type":21,"tag":322,"props":8818,"children":8819},{"style":348},[8820],{"type":26,"value":3783},{"type":21,"tag":322,"props":8822,"children":8823},{"style":359},[8824],{"type":26,"value":8825},", optimal iterations = ",{"type":21,"tag":322,"props":8827,"children":8828},{"style":348},[8829],{"type":26,"value":3773},{"type":21,"tag":322,"props":8831,"children":8832},{"style":329},[8833],{"type":26,"value":8834},"optimal",{"type":21,"tag":322,"props":8836,"children":8837},{"style":348},[8838],{"type":26,"value":3783},{"type":21,"tag":322,"props":8840,"children":8841},{"style":359},[8842],{"type":26,"value":3788},{"type":21,"tag":322,"props":8844,"children":8845},{"style":329},[8846],{"type":26,"value":3738},{"type":21,"tag":322,"props":8848,"children":8849},{"class":324,"line":4002},[8850],{"type":21,"tag":322,"props":8851,"children":8852},{"emptyLinePlaceholder":3662},[8853],{"type":26,"value":3665},{"type":21,"tag":322,"props":8855,"children":8856},{"class":324,"line":4044},[8857,8862,8866],{"type":21,"tag":322,"props":8858,"children":8859},{"style":329},[8860],{"type":26,"value":8861},"sim ",{"type":21,"tag":322,"props":8863,"children":8864},{"style":3638},[8865],{"type":26,"value":3678},{"type":21,"tag":322,"props":8867,"children":8868},{"style":329},[8869],{"type":26,"value":8870}," AerSimulator()\n",{"type":21,"tag":322,"props":8872,"children":8873},{"class":324,"line":5068},[8874,8878,8882],{"type":21,"tag":322,"props":8875,"children":8876},{"style":329},[8877],{"type":26,"value":7819},{"type":21,"tag":322,"props":8879,"children":8880},{"style":3638},[8881],{"type":26,"value":3678},{"type":21,"tag":322,"props":8883,"children":8884},{"style":329},[8885],{"type":26,"value":8886}," grover(n, marked, optimal)\n",{"type":21,"tag":322,"props":8888,"children":8889},{"class":324,"line":5087},[8890,8895,8899,8904,8908,8912,8917],{"type":21,"tag":322,"props":8891,"children":8892},{"style":329},[8893],{"type":26,"value":8894},"counts ",{"type":21,"tag":322,"props":8896,"children":8897},{"style":3638},[8898],{"type":26,"value":3678},{"type":21,"tag":322,"props":8900,"children":8901},{"style":329},[8902],{"type":26,"value":8903}," sim.run(transpile(qc, sim), ",{"type":21,"tag":322,"props":8905,"children":8906},{"style":3703},[8907],{"type":26,"value":4560},{"type":21,"tag":322,"props":8909,"children":8910},{"style":3638},[8911],{"type":26,"value":3678},{"type":21,"tag":322,"props":8913,"children":8914},{"style":348},[8915],{"type":26,"value":8916},"4096",{"type":21,"tag":322,"props":8918,"children":8919},{"style":329},[8920],{"type":26,"value":5719},{"type":21,"tag":322,"props":8922,"children":8923},{"class":324,"line":5095},[8924,8928,8932,8937],{"type":21,"tag":322,"props":8925,"children":8926},{"style":348},[8927],{"type":26,"value":3753},{"type":21,"tag":322,"props":8929,"children":8930},{"style":329},[8931],{"type":26,"value":3758},{"type":21,"tag":322,"props":8933,"children":8934},{"style":348},[8935],{"type":26,"value":8936},"sorted",{"type":21,"tag":322,"props":8938,"children":8939},{"style":329},[8940],{"type":26,"value":8941},"(counts.items()))\n",{"type":21,"tag":22,"props":8943,"children":8944},{},[8945],{"type":26,"value":8946},"Real output from this exact code (Qiskit 2.5, Aer 0.17):",{"type":21,"tag":312,"props":8948,"children":8950},{"code":8949},"N = 8, optimal iterations = 2\n[('000', 35), ('001', 34), ('010', 33), ('011', 27),\n ('100', 33), ('101', 3866), ('110', 35), ('111', 33)]\n",[8951],{"type":21,"tag":79,"props":8952,"children":8953},{"__ignoreMap":7},[8954],{"type":26,"value":8949},{"type":21,"tag":22,"props":8956,"children":8957},{},[8958,8960,8965],{"type":26,"value":8959},"94.4% on ",{"type":21,"tag":79,"props":8961,"children":8963},{"className":8962},[],[8964],{"type":26,"value":8241},{"type":26,"value":8966},", versus 12.5% by chance. Two oracle calls where classical search would need 4.5 on average.",{"type":21,"tag":59,"props":8968,"children":8970},{"id":8969},"the-mistake-more-iterations-make-it-worse",[8971],{"type":26,"value":8972},"The Mistake: More Iterations Make It Worse",{"type":21,"tag":22,"props":8974,"children":8975},{},[8976,8978,8983],{"type":26,"value":8977},"The intuition \"more amplification is better\" is wrong, and it's the single biggest trap in Grover's algorithm. Each iteration rotates the state by a ",{"type":21,"tag":71,"props":8979,"children":8980},{},[8981],{"type":26,"value":8982},"fixed",{"type":26,"value":8984}," angle θ where sin(θ\u002F2) = √(M\u002FN). After the optimal number of rounds you're aligned with the target. Keep going and you rotate straight past it, back down toward the unmarked states, and eventually back to where you started. Success probability oscillates as sin²((2k+1)·θ\u002F2).",{"type":21,"tag":22,"props":8986,"children":8987},{},[8988],{"type":26,"value":8989},"Loop the iteration count and watch. Measured counts over 4096 shots each:",{"type":21,"tag":312,"props":8991,"children":8993},{"code":8992},"iters=0  P(101)=0.127   \u003C- uniform, no better than guessing\niters=1  P(101)=0.779\niters=2  P(101)=0.944   \u003C- optimal\niters=3  P(101)=0.320   \u003C- overshooting\niters=4  P(101)=0.013   \u003C- worse than random guessing!\niters=5  P(101)=0.556\niters=6  P(101)=1.000   \u003C- period brings it back around\niters=7  P(101)=0.586\n",[8994],{"type":21,"tag":79,"props":8995,"children":8996},{"__ignoreMap":7},[8997],{"type":26,"value":8992},{"type":21,"tag":22,"props":8999,"children":9000},{},[9001,9003,9008],{"type":26,"value":9002},"Four iterations gives you a 1.3% success rate. You did ",{"type":21,"tag":51,"props":9004,"children":9005},{},[9006],{"type":26,"value":9007},"twice",{"type":26,"value":9009}," the optimal work and ended up nine times worse than doing nothing at all. This is not noise or a bug. It is exactly what the math says should happen, and it's the reason you must compute k in advance rather than \"running it a while\".",{"type":21,"tag":22,"props":9011,"children":9012},{},[9013,9015,9020],{"type":26,"value":9014},"The awkward corollary: computing k requires knowing M, the number of solutions. If you don't know how many marked states exist, you need ",{"type":21,"tag":71,"props":9016,"children":9017},{},[9018],{"type":26,"value":9019},"quantum counting",{"type":26,"value":9021}," (an amplitude-estimation routine) first, or you use randomized iteration counts, which costs you a constant factor. Textbook presentations tend to skip this.",{"type":21,"tag":59,"props":9023,"children":9025},{"id":9024},"what-the-speedup-buys-you",[9026],{"type":26,"value":9027},"What the Speedup Buys You",{"type":21,"tag":22,"props":9029,"children":9030},{},[9031],{"type":26,"value":9032},"Be precise about this, because a lot of writing on Grover is not.",{"type":21,"tag":22,"props":9034,"children":9035},{},[9036,9041,9043,9049],{"type":21,"tag":71,"props":9037,"children":9038},{},[9039],{"type":26,"value":9040},"It is quadratic, not exponential.",{"type":26,"value":9042}," √N versus N. ",{"type":21,"tag":29,"props":9044,"children":9046},{"href":9045},"\u002Fresearch\u002Fshor-factoring-1994",[9047],{"type":26,"value":9048},"Shor's factoring algorithm",{"type":26,"value":9050}," is exponential. That's a categorically different thing. A quadratic speedup is real and provably optimal (no quantum algorithm beats √N for unstructured search), but it is fragile: error correction's constant-factor overhead often eats it entirely, and that overhead is substantial.",{"type":21,"tag":22,"props":9052,"children":9053},{},[9054,9059],{"type":21,"tag":71,"props":9055,"children":9056},{},[9057],{"type":26,"value":9058},"It does not make NP-hard problems easy.",{"type":26,"value":9060}," Apply Grover to brute-force SAT and go from 2ⁿ to 2^(n\u002F2). That's a genuine improvement, and it is also still exponential. NP-hard problems remain NP-hard. Real SAT solvers use clause learning and structure that Grover, by definition, cannot exploit. That is why classical solvers routinely handle instances with millions of variables.",{"type":21,"tag":22,"props":9062,"children":9063},{},[9064,9069,9071,9076,9078,9083],{"type":21,"tag":71,"props":9065,"children":9066},{},[9067],{"type":26,"value":9068},"The \"database search\" framing is misleading.",{"type":26,"value":9070}," Grover needs an ",{"type":21,"tag":51,"props":9072,"children":9073},{},[9074],{"type":26,"value":9075},"oracle",{"type":26,"value":9077}," (a circuit that recognizes the answer), not a stored database. If you had N records in memory, you would first need to load them into quantum superposition, which costs O(N) and destroys the speedup outright. Grover is useful when the marked item is ",{"type":21,"tag":51,"props":9079,"children":9080},{},[9081],{"type":26,"value":9082},"defined by a computable property",{"type":26,"value":9084}," (this key decrypts the ciphertext, this input hashes to that digest) rather than looked up in a table.",{"type":21,"tag":22,"props":9086,"children":9087},{},[9088,9090,9096],{"type":26,"value":9089},"The real applications: cryptanalysis (Grover halves the effective key length of symmetric ciphers, which is exactly why AES-256 is recommended over AES-128 for post-quantum security), and as a subroutine inside larger algorithms via amplitude amplification. Grover's original 1996 paper is worth reading directly. See our ",{"type":21,"tag":29,"props":9091,"children":9093},{"href":9092},"\u002Fresearch\u002Fgrover-search-1996",[9094],{"type":26,"value":9095},"breakdown of the paper",{"type":26,"value":534},{"type":21,"tag":59,"props":9098,"children":9100},{"id":9099},"running-on-real-hardware",[9101],{"type":26,"value":9102},"Running on Real Hardware",{"type":21,"tag":22,"props":9104,"children":9105},{},[9106,9108,9113],{"type":26,"value":9107},"Moving to a QPU, the picture changes fast. Multi-controlled X gates are not native to any hardware. ",{"type":21,"tag":29,"props":9109,"children":9110},{"href":6192},[9111],{"type":26,"value":9112},"Transpilation",{"type":26,"value":9114}," decomposes each one into a cascade of CNOTs and single-qubit rotations, and the cost grows steeply with the number of controls. A 4-qubit MCX becomes 20+ two-qubit gates. Multiply that by two per Grover iteration, times k iterations, and depth explodes.",{"type":21,"tag":312,"props":9116,"children":9118},{"code":9117,"language":3628,"meta":7,"className":3626,"style":7},"# Check what your circuit costs before submitting\nqc = grover(3, \"101\", 2)\nprint(qc.decompose().count_ops())\n",[9119],{"type":21,"tag":79,"props":9120,"children":9121},{"__ignoreMap":7},[9122,9130,9171],{"type":21,"tag":322,"props":9123,"children":9124},{"class":324,"line":325},[9125],{"type":21,"tag":322,"props":9126,"children":9127},{"style":3888},[9128],{"type":26,"value":9129},"# Check what your circuit costs before submitting\n",{"type":21,"tag":322,"props":9131,"children":9132},{"class":324,"line":335},[9133,9137,9141,9146,9150,9154,9159,9163,9167],{"type":21,"tag":322,"props":9134,"children":9135},{"style":329},[9136],{"type":26,"value":7819},{"type":21,"tag":322,"props":9138,"children":9139},{"style":3638},[9140],{"type":26,"value":3678},{"type":21,"tag":322,"props":9142,"children":9143},{"style":329},[9144],{"type":26,"value":9145}," grover(",{"type":21,"tag":322,"props":9147,"children":9148},{"style":348},[9149],{"type":26,"value":6948},{"type":21,"tag":322,"props":9151,"children":9152},{"style":329},[9153],{"type":26,"value":367},{"type":21,"tag":322,"props":9155,"children":9156},{"style":359},[9157],{"type":26,"value":9158},"\"101\"",{"type":21,"tag":322,"props":9160,"children":9161},{"style":329},[9162],{"type":26,"value":367},{"type":21,"tag":322,"props":9164,"children":9165},{"style":348},[9166],{"type":26,"value":423},{"type":21,"tag":322,"props":9168,"children":9169},{"style":329},[9170],{"type":26,"value":3738},{"type":21,"tag":322,"props":9172,"children":9173},{"class":324,"line":344},[9174,9178],{"type":21,"tag":322,"props":9175,"children":9176},{"style":348},[9177],{"type":26,"value":3753},{"type":21,"tag":322,"props":9179,"children":9180},{"style":329},[9181],{"type":26,"value":9182},"(qc.decompose().count_ops())\n",{"type":21,"tag":22,"props":9184,"children":9185},{},[9186,9188,9194,9196,9200],{"type":26,"value":9187},"On ",{"type":21,"tag":29,"props":9189,"children":9191},{"href":9190},"\u002Fresearch\u002Fpreskill-nisq-2018",[9192],{"type":26,"value":9193},"NISQ devices",{"type":26,"value":9195},", two-qubit gate error sits around 0.5–1%. A few hundred entangling gates and your output is indistinguishable from uniform random. Published hardware demonstrations of Grover are almost all 2–3 qubits, and they are demonstrations, not useful computations. Add ",{"type":21,"tag":29,"props":9197,"children":9198},{"href":6145},[9199],{"type":26,"value":6148},{"type":26,"value":9201}," on top and you need enough samples to distinguish a modest peak from the floor.",{"type":21,"tag":22,"props":9203,"children":9204},{},[9205,9207,9212,9214,9219],{"type":26,"value":9206},"The practical advice: develop and debug on ",{"type":21,"tag":29,"props":9208,"children":9209},{"href":1005},[9210],{"type":26,"value":9211},"simulators",{"type":26,"value":9213},", where you verify correctness exactly. Only move to ",{"type":21,"tag":29,"props":9215,"children":9216},{"href":806},[9217],{"type":26,"value":9218},"real hardware",{"type":26,"value":9220}," when you understand what your circuit costs in native gates, and expect degraded results.",{"type":21,"tag":59,"props":9222,"children":9224},{"id":9223},"next-steps",[9225],{"type":26,"value":9226},"Next Steps",{"type":21,"tag":3556,"props":9228,"children":9229},{},[9230,9243,9255,9276],{"type":21,"tag":3560,"props":9231,"children":9232},{},[9233,9241],{"type":21,"tag":71,"props":9234,"children":9235},{},[9236],{"type":21,"tag":29,"props":9237,"children":9238},{"href":7528},[9239],{"type":26,"value":9240},"Qiskit SDK guide",{"type":26,"value":9242},": full setup, backends, and transpiler options",{"type":21,"tag":3560,"props":9244,"children":9245},{},[9246,9253],{"type":21,"tag":71,"props":9247,"children":9248},{},[9249],{"type":21,"tag":29,"props":9250,"children":9251},{"href":31},[9252],{"type":26,"value":662},{"type":26,"value":9254},": the other big near-term algorithm, applied to Max-Cut",{"type":21,"tag":3560,"props":9256,"children":9257},{},[9258,9266,9268,9274],{"type":21,"tag":71,"props":9259,"children":9260},{},[9261],{"type":21,"tag":29,"props":9262,"children":9263},{"href":814},[9264],{"type":26,"value":9265},"Compare the SDKs",{"type":26,"value":9267},": the same Grover circuit in ",{"type":21,"tag":29,"props":9269,"children":9271},{"href":9270},"\u002Fsdks\u002Fcirq",[9272],{"type":26,"value":9273},"Cirq",{"type":26,"value":9275}," or PennyLane",{"type":21,"tag":3560,"props":9277,"children":9278},{},[9279,9287,9288,9294,9295,9301],{"type":21,"tag":71,"props":9280,"children":9281},{},[9282],{"type":21,"tag":29,"props":9283,"children":9284},{"href":4400},[9285],{"type":26,"value":9286},"Glossary",{"type":26,"value":418},{"type":21,"tag":29,"props":9289,"children":9291},{"href":9290},"\u002Fglossary\u002Fsuperposition",[9292],{"type":26,"value":9293},"superposition",{"type":26,"value":367},{"type":21,"tag":29,"props":9296,"children":9298},{"href":9297},"\u002Fglossary\u002Fquantum-advantage",[9299],{"type":26,"value":9300},"quantum advantage",{"type":26,"value":9302},", and the rest of the vocabulary",{"type":21,"tag":703,"props":9304,"children":9305},{},[9306],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":9308},[9309,9310,9311,9312,9313,9314,9315,9316,9317,9318],{"id":7607,"depth":335,"text":7610},{"id":7635,"depth":335,"text":7638},{"id":7681,"depth":335,"text":7684},{"id":7865,"depth":335,"text":7868},{"id":8246,"depth":335,"text":8249},{"id":8492,"depth":335,"text":8495},{"id":8969,"depth":335,"text":8972},{"id":9024,"depth":335,"text":9027},{"id":9099,"depth":335,"text":9102},{"id":9223,"depth":335,"text":9226},"content:blog:grover-algorithm-tutorial.md","blog\u002Fgrover-algorithm-tutorial.md","blog\u002Fgrover-algorithm-tutorial",{"_path":751,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":9323,"description":9324,"date":3500,"author":11,"tags":9325,"readingTime":496,"body":9327,"_type":715,"_id":9778,"_source":717,"_file":9779,"_stem":9780,"_extension":720},"Logical Qubits and Fault Tolerance: What the Qubit Count Doesn't Tell You","A physical qubit is hardware. A logical qubit is an error-protected qubit built out of many of them. Understanding the difference is the fastest way to see through quantum computing headlines.",[729,730,9326],"QPU",{"type":18,"children":9328,"toc":9768},[9329,9334,9346,9352,9379,9393,9420,9426,9438,9443,9456,9475,9481,9486,9497,9510,9515,9520,9526,9538,9543,9562,9567,9573,9578,9588,9598,9621,9627,9632,9642,9652,9662,9672,9677,9683,9688,9693,9719,9738,9744,9756],{"type":21,"tag":22,"props":9330,"children":9331},{},[9332],{"type":26,"value":9333},"Every few months a press release announces a new record qubit count. Almost none of them tell you the number that decides whether the machine does anything: the error rate.",{"type":21,"tag":22,"props":9335,"children":9336},{},[9337,9339,9344],{"type":26,"value":9338},"This is not a pedantic complaint. The relationship between qubit count and computational power in quantum computing is not merely noisy. It sometimes inverts. Under the right conditions, a machine with more qubits is ",{"type":21,"tag":51,"props":9340,"children":9341},{},[9342],{"type":26,"value":9343},"worse",{"type":26,"value":9345}," than one with fewer. Understanding why requires one distinction, and once you have it, most quantum computing news becomes far easier to read.",{"type":21,"tag":59,"props":9347,"children":9349},{"id":9348},"two-things-called-qubit",[9350],{"type":26,"value":9351},"Two things called \"qubit\"",{"type":21,"tag":22,"props":9353,"children":9354},{},[9355,9357,9362,9364,9370,9372,9377],{"type":26,"value":9356},"A ",{"type":21,"tag":71,"props":9358,"children":9359},{},[9360],{"type":26,"value":9361},"physical qubit",{"type":26,"value":9363}," is a piece of hardware: a superconducting circuit on a chip, a trapped ion held in an electromagnetic field, a neutral atom in an optical tweezer. It is a real object subject to real physics, which means it drifts, absorbs stray energy, and loses its quantum state on a timescale measured in microseconds to seconds. That decay is ",{"type":21,"tag":29,"props":9365,"children":9367},{"href":9366},"\u002Fglossary\u002Fdecoherence",[9368],{"type":26,"value":9369},"decoherence",{"type":26,"value":9371},", and the ",{"type":21,"tag":29,"props":9373,"children":9374},{"href":6171},[9375],{"type":26,"value":9376},"T1 and T2 times",{"type":26,"value":9378}," that quantify it are among the few hardware specs worth reading closely.",{"type":21,"tag":22,"props":9380,"children":9381},{},[9382,9383,9391],{"type":26,"value":9356},{"type":21,"tag":71,"props":9384,"children":9385},{},[9386],{"type":21,"tag":29,"props":9387,"children":9388},{"href":1983},[9389],{"type":26,"value":9390},"logical qubit",{"type":26,"value":9392}," is not hardware. It is a qubit's worth of quantum information spread across many physical qubits, arranged so that errors on individual members are detected and undone without ever measuring, and thereby destroying, the encoded state. IonQ describes logical qubits as software-defined, and that framing is a good one: the logical qubit is a construction that hardware supports rather than a thing you point to.",{"type":21,"tag":22,"props":9394,"children":9395},{},[9396,9398,9403,9405,9410,9412,9418],{"type":26,"value":9397},"The technique is ",{"type":21,"tag":29,"props":9399,"children":9400},{"href":1526},[9401],{"type":26,"value":9402},"quantum error correction",{"type":26,"value":9404},", and our ",{"type":21,"tag":29,"props":9406,"children":9407},{"href":854},[9408],{"type":26,"value":9409},"companion post on QEC",{"type":26,"value":9411}," covers the mechanics: parity checks, ",{"type":21,"tag":29,"props":9413,"children":9415},{"href":9414},"\u002Fglossary\u002Fancilla-qubit",[9416],{"type":26,"value":9417},"ancilla qubits",{"type":26,"value":9419},", the surface code lattice. This post is about the accounting: what encoding costs, when it pays off, and what else fault tolerance demands beyond the encoding itself.",{"type":21,"tag":59,"props":9421,"children":9423},{"id":9422},"why-how-many-qubits-is-nearly-meaningless",[9424],{"type":26,"value":9425},"Why \"how many qubits?\" is nearly meaningless",{"type":21,"tag":22,"props":9427,"children":9428},{},[9429,9431,9436],{"type":26,"value":9430},"Here is the uncomfortable arithmetic. If your physical error rate is above the threshold of your error-correcting code, encoding does not help. The extra qubits and the extra gates needed to perform parity checks each introduce errors of their own. Add more of them and you add more noise than you remove. A headline of 1,000 physical qubits with mediocre fidelity is compatible with ",{"type":21,"tag":71,"props":9432,"children":9433},{},[9434],{"type":26,"value":9435},"zero",{"type":26,"value":9437}," logical qubits, not few, zero.",{"type":21,"tag":22,"props":9439,"children":9440},{},[9441],{"type":26,"value":9442},"Meanwhile a smaller, cleaner machine is sometimes strictly more useful. IonQ argues this point aggressively, claiming that a system of 100 physical qubits at 99.99% two-qubit gate fidelity would likely outperform a 10,000-qubit system of lower-quality qubits encoding 100 logical ones: on overhead, gate speed, universality, and energy. That's a vendor making a case for its own architecture and should be read as such, but the underlying logic is sound and widely accepted: quality compounds in a way that quantity does not.",{"type":21,"tag":22,"props":9444,"children":9445},{},[9446,9448,9454],{"type":26,"value":9447},"The compounding is the key mechanism. In a code that corrects one error, roughly speaking, halving the physical error rate quarters the logical error rate. In a code correcting two errors the same improvement gives about an eight-fold gain. IonQ makes this multiplicative argument explicitly, and it explains why hardware teams chase ",{"type":21,"tag":29,"props":9449,"children":9451},{"href":9450},"\u002Fglossary\u002Ffidelity",[9452],{"type":26,"value":9453},"fidelity",{"type":26,"value":9455}," improvements that look small in isolation. A 2x hardware win is a 4x or 8x win after encoding.",{"type":21,"tag":22,"props":9457,"children":9458},{},[9459,9461,9466,9468,9473],{"type":26,"value":9460},"This is also why single-number benchmarks keep failing the field. ",{"type":21,"tag":29,"props":9462,"children":9463},{"href":3578},[9464],{"type":26,"value":9465},"Quantum Volume",{"type":26,"value":9467},", algorithmic qubits, gate counts: each captures a slice and hides the rest. We wrote about that measurement problem ",{"type":21,"tag":29,"props":9469,"children":9470},{"href":1274},[9471],{"type":26,"value":9472},"in the context of EO 14413",{"type":26,"value":9474},", and it applies with full force here.",{"type":21,"tag":59,"props":9476,"children":9478},{"id":9477},"the-overhead-is-the-whole-story",[9479],{"type":26,"value":9480},"The overhead is the whole story",{"type":21,"tag":22,"props":9482,"children":9483},{},[9484],{"type":26,"value":9485},"How many physical qubits does one logical qubit cost? The honest answer is that it depends on two numbers you have to state together: your physical error rate and the logical error rate you're targeting.",{"type":21,"tag":22,"props":9487,"children":9488},{},[9489,9495],{"type":21,"tag":29,"props":9490,"children":9492},{"href":9491},"\u002Fresearch\u002Fshor-error-correction-1995",[9493],{"type":26,"value":9494},"Shor's 1995 code",{"type":26,"value":9496},", the first quantum error-correcting code ever written down, used nine physical qubits to protect one logical qubit against an arbitrary single-qubit error. That was a proof that the thing was possible at all. Before it, many physicists believed no-cloning made quantum computing hopeless in principle.",{"type":21,"tag":22,"props":9498,"children":9499},{},[9500,9502,9508],{"type":26,"value":9501},"Nine turned out to be optimistic for practical machines. ",{"type":21,"tag":29,"props":9503,"children":9505},{"href":9504},"\u002Fresearch\u002Fkitaev-anyons-1997",[9506],{"type":26,"value":9507},"Kitaev's 1997 work on anyons",{"type":26,"value":9509}," introduced topological codes, including the surface code, which stores information in global properties of a 2D lattice rather than in any individual site. Its great virtue is that every parity check involves only neighbouring qubits, which maps cleanly onto flat chips, and it tolerates physical error rates around 1%, high enough that real hardware plausibly reaches it.",{"type":21,"tag":22,"props":9511,"children":9512},{},[9513],{"type":26,"value":9514},"The price is scale. Surface code overhead grows with code distance, and realistic estimates for running a cryptographically relevant algorithm land at hundreds to a few thousand physical qubits per logical qubit, with total system requirements often quoted in the millions. Push your physical error rate down and that ratio falls fast. Let it drift up toward threshold and the ratio explodes. Overhead is not a fixed constant of the technology. It is a function of how good your hardware is.",{"type":21,"tag":22,"props":9516,"children":9517},{},[9518],{"type":26,"value":9519},"Newer code families are attacking the ratio directly. IonQ has promoted a bivariate bicycle variant it calls BB5, claiming an idle logical error rate around 5x10⁻⁵ using 50 physical qubits, roughly four times smaller than standard BB codes. Vendor-reported figures like that are worth tracking but not worth treating as settled until independently reproduced.",{"type":21,"tag":59,"props":9521,"children":9523},{"id":9522},"the-threshold-theorem-and-why-2024-mattered",[9524],{"type":26,"value":9525},"The threshold theorem, and why 2024 mattered",{"type":21,"tag":22,"props":9527,"children":9528},{},[9529,9531,9536],{"type":26,"value":9530},"The theoretical foundation under all of this is the ",{"type":21,"tag":71,"props":9532,"children":9533},{},[9534],{"type":26,"value":9535},"threshold theorem",{"type":26,"value":9537},". It says there exists a critical physical error rate. Below it, increasing the size of your code suppresses logical errors exponentially. You make them as small as you like by spending more qubits. Above it, the opposite: bigger codes are worse codes.",{"type":21,"tag":22,"props":9539,"children":9540},{},[9541],{"type":26,"value":9542},"For decades this was a theorem without an experiment. Every roadmap in the industry assumed the crossing was achievable. None had demonstrated it.",{"type":21,"tag":22,"props":9544,"children":9545},{},[9546,9548,9553,9555,9560],{"type":26,"value":9547},"That changed with ",{"type":21,"tag":29,"props":9549,"children":9550},{"href":4184},[9551],{"type":26,"value":9552},"Google's below-threshold result in 2024",{"type":26,"value":9554},". Running surface codes at distances 3, 5, and 7 on its 105-qubit Willow processor, the team observed each increase in code distance roughly ",{"type":21,"tag":71,"props":9556,"children":9557},{},[9558],{"type":26,"value":9559},"halving",{"type":26,"value":9561}," the logical error rate, the direction fault tolerance requires. Critically, the encoded logical qubit outlived the best individual physical qubit on the chip, which is the concrete test of whether error correction is a net win rather than an expensive way to add noise.",{"type":21,"tag":22,"props":9563,"children":9564},{},[9565],{"type":26,"value":9566},"It's hard to overstate the significance. That experiment converted large-scale quantum computing from an open physics question into a scaling and engineering problem. Engineering problems are hard, but they are a different category of hard.",{"type":21,"tag":59,"props":9568,"children":9570},{"id":9569},"different-hardware-different-arithmetic",[9571],{"type":26,"value":9572},"Different hardware, different arithmetic",{"type":21,"tag":22,"props":9574,"children":9575},{},[9576],{"type":26,"value":9577},"The threshold is not a single universal number. It depends on the code, and which codes are practical depends on your hardware's connectivity. This is where modality differences stop being trivia.",{"type":21,"tag":22,"props":9579,"children":9580},{},[9581,9586],{"type":21,"tag":71,"props":9582,"children":9583},{},[9584],{"type":26,"value":9585},"Superconducting",{"type":26,"value":9587}," processors offer fast gates (nanoseconds) and mature fabrication, but qubits interact only with their planar neighbours and fidelities are typically lower. That planar constraint is precisely what the surface code was designed around, which is why superconducting roadmaps are built on it.",{"type":21,"tag":22,"props":9589,"children":9590},{},[9591,9596],{"type":21,"tag":71,"props":9592,"children":9593},{},[9594],{"type":26,"value":9595},"Trapped ions",{"type":26,"value":9597}," invert the trade: gates are far slower (microseconds to milliseconds), but fidelities are the highest of any modality and connectivity is effectively all-to-all: any ion in a chain is entangled with any other without a chain of intervening swap operations. IonQ argues this connectivity is a structural advantage for error correction, because codes requiring non-local checks become implementable rather than prohibitively expensive, and it has claimed 99.99% physical two-qubit gate fidelity via its Oxford Ionics acquisition. Again: vendor claim, vendor benchmark conditions.",{"type":21,"tag":22,"props":9599,"children":9600},{},[9601,9603,9607,9608,9613,9615,9619],{"type":26,"value":9602},"Neither is obviously winning. Our ",{"type":21,"tag":29,"props":9604,"children":9605},{"href":806},[9606],{"type":26,"value":1706},{"type":26,"value":236},{"type":21,"tag":29,"props":9609,"children":9610},{"href":814},[9611],{"type":26,"value":9612},"modality comparison",{"type":26,"value":9614}," go into the specifics, and the broader ",{"type":21,"tag":29,"props":9616,"children":9617},{"href":862},[9618],{"type":26,"value":1452},{"type":26,"value":9620}," tracks who is betting on what. The practical takeaway for anyone learning is that connectivity and fidelity, not qubit count, are the specs that determine which error-correction strategies a machine even attempts.",{"type":21,"tag":59,"props":9622,"children":9624},{"id":9623},"encoding-is-necessary-not-sufficient",[9625],{"type":26,"value":9626},"Encoding is necessary, not sufficient",{"type":21,"tag":22,"props":9628,"children":9629},{},[9630],{"type":26,"value":9631},"A subtlety that gets lost in coverage: storing a logical qubit is the easy part of fault tolerance. Computing on one is harder. A genuinely fault-tolerant machine needs all of the following:",{"type":21,"tag":22,"props":9633,"children":9634},{},[9635,9640],{"type":21,"tag":71,"props":9636,"children":9637},{},[9638],{"type":26,"value":9639},"Fault-tolerant gate operations.",{"type":26,"value":9641}," Logical gates must be implemented so that a single physical fault cannot propagate into an uncorrectable logical error. Some gates are cheap in a given code. Others are not.",{"type":21,"tag":22,"props":9643,"children":9644},{},[9645,9650],{"type":21,"tag":71,"props":9646,"children":9647},{},[9648],{"type":26,"value":9649},"Continuous syndrome extraction.",{"type":26,"value":9651}," Parity checks run constantly, in rounds, throughout the computation. The measurement circuits are themselves noisy, so the scheme has to tolerate faults in its own error detection.",{"type":21,"tag":22,"props":9653,"children":9654},{},[9655,9660],{"type":21,"tag":71,"props":9656,"children":9657},{},[9658],{"type":26,"value":9659},"Magic state distillation.",{"type":26,"value":9661}," Surface codes give you Clifford gates relatively cheaply, but Clifford gates alone are classically simulable. Universality needs a non-Clifford gate, typically T, and those are produced by distilling noisy \"magic states\" into clean ones. Distillation factories consume a large fraction of the total qubit budget in realistic architectures.",{"type":21,"tag":22,"props":9663,"children":9664},{},[9665,9670],{"type":21,"tag":71,"props":9666,"children":9667},{},[9668],{"type":26,"value":9669},"Real-time decoding.",{"type":26,"value":9671}," Syndrome data must be interpreted and corrections applied faster than errors accumulate. This is a classical computing problem, running at microsecond latency alongside the QPU, and it's a serious engineering constraint in its own right.",{"type":21,"tag":22,"props":9673,"children":9674},{},[9675],{"type":26,"value":9676},"IonQ's framing here is useful regardless of the vendor context: a logical qubit should be characterised by several attributes together: overhead, idle logical error rate, logical gate fidelity, logical gate speed, and gate-set universality, rather than counted. A logical qubit that is stored but not usefully operated on is not much of a logical qubit.",{"type":21,"tag":59,"props":9678,"children":9680},{"id":9679},"where-the-field-stands",[9681],{"type":26,"value":9682},"Where the field stands",{"type":21,"tag":22,"props":9684,"children":9685},{},[9686],{"type":26,"value":9687},"Honest summary as of mid-2026: logical qubits are real, demonstrated, and few.",{"type":21,"tag":22,"props":9689,"children":9690},{},[9691],{"type":26,"value":9692},"Multiple groups have encoded them. Google has shown error suppression scaling in the right direction. Trapped-ion and neutral-atom teams have run algorithms on small numbers of encoded qubits. These are genuine milestones, not marketing.",{"type":21,"tag":22,"props":9694,"children":9695},{},[9696,9698,9703,9705,9710,9712,9717],{"type":26,"value":9697},"But useful fault-tolerant computation needs hundreds to thousands of logical qubits executing millions of logical operations, and that means physical qubit counts several orders of magnitude beyond anything running today, with error rates comfortably below threshold across the whole device rather than on the best-behaved corner of a chip. Most working hardware remains firmly in the ",{"type":21,"tag":29,"props":9699,"children":9700},{"href":263},[9701],{"type":26,"value":9702},"NISQ regime",{"type":26,"value":9704}," that ",{"type":21,"tag":29,"props":9706,"children":9707},{"href":9190},[9708],{"type":26,"value":9709},"Preskill named in 2018",{"type":26,"value":9711},", where ",{"type":21,"tag":29,"props":9713,"children":9714},{"href":4484},[9715],{"type":26,"value":9716},"error mitigation",{"type":26,"value":9718}," (statistical post-processing rather than true correction) is the practical tool.",{"type":21,"tag":22,"props":9720,"children":9721},{},[9722,9724,9729,9731,9736],{"type":26,"value":9723},"None of which means waiting around. The abstractions transfer: circuits, gates, measurement, and noise behave the same way whether you're on a simulator or a fault-tolerant machine a decade out. You start on ",{"type":21,"tag":29,"props":9725,"children":9726},{"href":1005},[9727],{"type":26,"value":9728},"free simulators",{"type":26,"value":9730}," or real QPUs today, and the ",{"type":21,"tag":29,"props":9732,"children":9733},{"href":4314},[9734],{"type":26,"value":9735},"courses page",{"type":26,"value":9737}," collects structured routes in.",{"type":21,"tag":59,"props":9739,"children":9741},{"id":9740},"the-one-habit-worth-forming",[9742],{"type":26,"value":9743},"The one habit worth forming",{"type":21,"tag":22,"props":9745,"children":9746},{},[9747,9749,9754],{"type":26,"value":9748},"When you next see a qubit-count headline, ask three questions: what is the two-qubit gate fidelity, is it below the relevant threshold, and how many ",{"type":21,"tag":51,"props":9750,"children":9751},{},[9752],{"type":26,"value":9753},"logical",{"type":26,"value":9755}," qubits does that imply?",{"type":21,"tag":22,"props":9757,"children":9758},{},[9759,9761,9766],{"type":26,"value":9760},"Frequently the answer to the third is zero, and the article won't have mentioned it. Learning to notice that gap is most of what separates informed reading from press-release reading. The ",{"type":21,"tag":29,"props":9762,"children":9763},{"href":4400},[9764],{"type":26,"value":9765},"glossary",{"type":26,"value":9767}," is a decent place to build the vocabulary for it.",{"title":7,"searchDepth":335,"depth":335,"links":9769},[9770,9771,9772,9773,9774,9775,9776,9777],{"id":9348,"depth":335,"text":9351},{"id":9422,"depth":335,"text":9425},{"id":9477,"depth":335,"text":9480},{"id":9522,"depth":335,"text":9525},{"id":9569,"depth":335,"text":9572},{"id":9623,"depth":335,"text":9626},{"id":9679,"depth":335,"text":9682},{"id":9740,"depth":335,"text":9743},"content:blog:logical-qubits-fault-tolerance-explained.md","blog\u002Flogical-qubits-fault-tolerance-explained.md","blog\u002Flogical-qubits-fault-tolerance-explained",{"_path":2343,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":9782,"description":9783,"date":3500,"author":11,"tags":9784,"readingTime":496,"body":9788,"_type":715,"_id":10886,"_source":717,"_file":10887,"_stem":10888,"_extension":720},"The Post-Quantum Migration Clock: Real Deadlines and How to Plan Against Them","NIST says RSA and ECC are deprecated after 2030 and disallowed after 2035. NSA's CNSA 2.0 moves faster. Here's what the published timelines require, and how to build a migration plan that survives them.",[9785,9786,9787],"Post-Quantum","Security","Cryptography",{"type":18,"children":9789,"toc":10876},[9790,9810,9823,9829,9834,9944,9949,9968,9974,9979,10009,10019,10029,10041,10046,10052,10072,10077,10103,10114,10120,10125,10168,10174,10179,10599,10611,10617,10629,10641,10691,10696,10702,10707,10816,10822,10834,10839,10872],{"type":21,"tag":22,"props":9791,"children":9792},{},[9793,9795,9801,9803,9808],{"type":26,"value":9794},"Most discussions of ",{"type":21,"tag":29,"props":9796,"children":9798},{"href":9797},"\u002Fglossary\u002Fpost-quantum-cryptography",[9799],{"type":26,"value":9800},"post-quantum cryptography",{"type":26,"value":9802}," stall at the same place: ",{"type":21,"tag":51,"props":9804,"children":9805},{},[9806],{"type":26,"value":9807},"when do we have to do this?",{"type":26,"value":9809}," The honest answer used to be \"nobody knows.\" That's no longer true. Not because anyone predicts when a cryptographically relevant quantum computer arrives, but because regulators stopped waiting for that prediction and published dates anyway.",{"type":21,"tag":22,"props":9811,"children":9812},{},[9813,9815,9821],{"type":26,"value":9814},"If you already understand what PQC is and which algorithms replace which, our ",{"type":21,"tag":29,"props":9816,"children":9818},{"href":9817},"\u002Fblog\u002Fpost-quantum-cryptography-guide",[9819],{"type":26,"value":9820},"guide to post-quantum cryptography",{"type":26,"value":9822}," covers that ground. This post is about the calendar and the project plan.",{"type":21,"tag":59,"props":9824,"children":9826},{"id":9825},"the-standards-are-settled",[9827],{"type":26,"value":9828},"The standards are settled",{"type":21,"tag":22,"props":9830,"children":9831},{},[9832],{"type":26,"value":9833},"NIST finalized three post-quantum standards on August 13, 2024, after an eight-year competition:",{"type":21,"tag":4503,"props":9835,"children":9836},{},[9837,9863],{"type":21,"tag":4507,"props":9838,"children":9839},{},[9840],{"type":21,"tag":4511,"props":9841,"children":9842},{},[9843,9848,9853,9858],{"type":21,"tag":4515,"props":9844,"children":9845},{},[9846],{"type":26,"value":9847},"Standard",{"type":21,"tag":4515,"props":9849,"children":9850},{},[9851],{"type":26,"value":9852},"Algorithm",{"type":21,"tag":4515,"props":9854,"children":9855},{},[9856],{"type":26,"value":9857},"Formerly",{"type":21,"tag":4515,"props":9859,"children":9860},{},[9861],{"type":26,"value":9862},"Replaces",{"type":21,"tag":4529,"props":9864,"children":9865},{},[9866,9892,9918],{"type":21,"tag":4511,"props":9867,"children":9868},{},[9869,9877,9882,9887],{"type":21,"tag":4536,"props":9870,"children":9871},{},[9872],{"type":21,"tag":71,"props":9873,"children":9874},{},[9875],{"type":26,"value":9876},"FIPS 203",{"type":21,"tag":4536,"props":9878,"children":9879},{},[9880],{"type":26,"value":9881},"ML-KEM",{"type":21,"tag":4536,"props":9883,"children":9884},{},[9885],{"type":26,"value":9886},"CRYSTALS-Kyber",{"type":21,"tag":4536,"props":9888,"children":9889},{},[9890],{"type":26,"value":9891},"RSA key transport, ECDH",{"type":21,"tag":4511,"props":9893,"children":9894},{},[9895,9903,9908,9913],{"type":21,"tag":4536,"props":9896,"children":9897},{},[9898],{"type":21,"tag":71,"props":9899,"children":9900},{},[9901],{"type":26,"value":9902},"FIPS 204",{"type":21,"tag":4536,"props":9904,"children":9905},{},[9906],{"type":26,"value":9907},"ML-DSA",{"type":21,"tag":4536,"props":9909,"children":9910},{},[9911],{"type":26,"value":9912},"CRYSTALS-Dilithium",{"type":21,"tag":4536,"props":9914,"children":9915},{},[9916],{"type":26,"value":9917},"ECDSA, RSA-PSS signatures",{"type":21,"tag":4511,"props":9919,"children":9920},{},[9921,9929,9934,9939],{"type":21,"tag":4536,"props":9922,"children":9923},{},[9924],{"type":21,"tag":71,"props":9925,"children":9926},{},[9927],{"type":26,"value":9928},"FIPS 205",{"type":21,"tag":4536,"props":9930,"children":9931},{},[9932],{"type":26,"value":9933},"SLH-DSA",{"type":21,"tag":4536,"props":9935,"children":9936},{},[9937],{"type":26,"value":9938},"SPHINCS+",{"type":21,"tag":4536,"props":9940,"children":9941},{},[9942],{"type":26,"value":9943},"Conservative hash-based signing",{"type":21,"tag":22,"props":9945,"children":9946},{},[9947],{"type":26,"value":9948},"Use the FIPS names in procurement documents and internal specs. \"Kyber\" and \"Dilithium\" refer to the competition submissions, which differ in small but real ways from the standardized versions. A vendor claiming \"Kyber support\" does or doesn't mean FIPS 203 compliance, and that ambiguity has already caused interop problems.",{"type":21,"tag":22,"props":9950,"children":9951},{},[9952,9954,9959,9961,9966],{"type":26,"value":9953},"Two more are in flight. NIST selected ",{"type":21,"tag":71,"props":9955,"children":9956},{},[9957],{"type":26,"value":9958},"HQC",{"type":26,"value":9960}," in March 2025 as a code-based backup KEM with different mathematical foundations from ML-KEM, with a draft standard expected around 2026 and finalization targeted for 2027. ",{"type":21,"tag":71,"props":9962,"children":9963},{},[9964],{"type":26,"value":9965},"FN-DSA",{"type":26,"value":9967}," (FALCON) remains slated for standardization as a compact-signature option. Neither should hold up your migration. ML-KEM and ML-DSA are the workhorses.",{"type":21,"tag":59,"props":9969,"children":9971},{"id":9970},"the-deadlines-that-exist-on-paper",[9972],{"type":26,"value":9973},"The deadlines that exist on paper",{"type":21,"tag":22,"props":9975,"children":9976},{},[9977],{"type":26,"value":9978},"Three separate sets of dates matter, and they don't line up.",{"type":21,"tag":22,"props":9980,"children":9981},{},[9982,9987,9989,9994,9996,10001,10002,10007],{"type":21,"tag":71,"props":9983,"children":9984},{},[9985],{"type":26,"value":9986},"NIST IR 8547",{"type":26,"value":9988}," (",{"type":21,"tag":51,"props":9990,"children":9991},{},[9992],{"type":26,"value":9993},"Transition to Post-Quantum Cryptography Standards",{"type":26,"value":9995},", initial public draft November 2024) is the broadest. It proposes that RSA, ECDSA, EdDSA, ECDH, DSA, and finite-field Diffie-Hellman become ",{"type":21,"tag":71,"props":9997,"children":9998},{},[9999],{"type":26,"value":10000},"deprecated after 2030",{"type":26,"value":236},{"type":21,"tag":71,"props":10003,"children":10004},{},[10005],{"type":26,"value":10006},"disallowed after 2035",{"type":26,"value":10008},". Deprecated means continued use requires the data owner to document a risk acceptance. Disallowed means the option to accept that risk goes away.",{"type":21,"tag":22,"props":10010,"children":10011},{},[10012,10017],{"type":21,"tag":71,"props":10013,"children":10014},{},[10015],{"type":26,"value":10016},"NSA's CNSA 2.0",{"type":26,"value":10018}," applies to National Security Systems and moves faster, with a staged schedule by product category. As published and updated through May 2025, software and firmware signing is the leading edge (exclusive use of CNSA 2.0 algorithms from January 1, 2027), with networking equipment following around 2030, and operating systems, custom applications, and cloud services around 2033. Full quantum resistance across NSS is expected by 2035.",{"type":21,"tag":22,"props":10020,"children":10021},{},[10022,10027],{"type":21,"tag":71,"props":10023,"children":10024},{},[10025],{"type":26,"value":10026},"OMB M-23-02",{"type":26,"value":10028}," (November 2022) is the one that already bit. It directed federal agencies to inventory cryptography on prioritized systems, produce funding estimates, and report annually, with full federal migration targeted for 2035.",{"type":21,"tag":22,"props":10030,"children":10031},{},[10032,10034,10039],{"type":26,"value":10033},"Two things follow from this. First, if you sell to the US federal government, especially anything involving signed firmware, your effective deadline is ",{"type":21,"tag":71,"props":10035,"children":10036},{},[10037],{"type":26,"value":10038},"2027, not 2035",{"type":26,"value":10040},". Second, even for purely commercial work, NIST's deprecation calendar becomes the de facto compliance baseline the moment auditors and insurers start citing it, which historically happens well before the date itself.",{"type":21,"tag":22,"props":10042,"children":10043},{},[10044],{"type":26,"value":10045},"Note that none of these dates are predictions about quantum hardware. They are engineering schedules working backward from the migration itself taking a decade.",{"type":21,"tag":59,"props":10047,"children":10049},{"id":10048},"why-the-threat-model-has-no-start-date",[10050],{"type":26,"value":10051},"Why the threat model has no start date",{"type":21,"tag":22,"props":10053,"children":10054},{},[10055,10057,10062,10064,10070],{"type":26,"value":10056},"The reason a 2030 deadline is defensible even though nobody expects a code-breaking quantum computer by 2030 is ",{"type":21,"tag":71,"props":10058,"children":10059},{},[10060],{"type":26,"value":10061},"harvest now, decrypt later",{"type":26,"value":10063},". An adversary records encrypted traffic today and decrypts it whenever ",{"type":21,"tag":29,"props":10065,"children":10067},{"href":10066},"\u002Fglossary\u002Fshor-algorithm",[10068],{"type":26,"value":10069},"Shor's algorithm",{"type":26,"value":10071}," becomes runnable at scale. Any data whose confidentiality must outlive that moment is already exposed.",{"type":21,"tag":22,"props":10073,"children":10074},{},[10075],{"type":26,"value":10076},"The practical calculation is Mosca's inequality: if the time your data must stay secret plus the time your migration takes exceeds the time until a capable quantum computer exists, you are already late. For medical records with a 50-year sensitivity horizon and a five-year migration, you needed to start before the standards existed.",{"type":21,"tag":22,"props":10078,"children":10079},{},[10080,10082,10087,10089,10093,10095,10101],{"type":26,"value":10081},"What makes this hard to plan against is that the hardware estimate keeps moving. ",{"type":21,"tag":29,"props":10083,"children":10084},{"href":9045},[10085],{"type":26,"value":10086},"Shor's 1994 paper",{"type":26,"value":10088}," established the algorithm. The resource question (how many ",{"type":21,"tag":29,"props":10090,"children":10091},{"href":1983},[10092],{"type":26,"value":1986},{"type":26,"value":10094}," and how many gates to break RSA-2048) has been re-estimated downward repeatedly as compilation techniques improved. Recent work ",{"type":21,"tag":29,"props":10096,"children":10098},{"href":10097},"\u002Fresearch\u002Fshor-lean-formalization-2026",[10099],{"type":26,"value":10100},"machine-checking those estimates in Lean",{"type":26,"value":10102}," is valuable precisely because it removes hand-waving from a number that security planning depends on. The estimates remain far beyond current devices, but \"far\" has been shrinking, and the direction of travel is one-way.",{"type":21,"tag":22,"props":10104,"children":10105},{},[10106,10108,10112],{"type":26,"value":10107},"For the record: symmetric cryptography is fine. ",{"type":21,"tag":29,"props":10109,"children":10110},{"href":7594},[10111],{"type":26,"value":7597},{"type":26,"value":10113}," gives only a quadratic speedup, so AES-256 retains 128-bit effective security. The migration is an asymmetric-crypto migration.",{"type":21,"tag":59,"props":10115,"children":10117},{"id":10116},"why-this-takes-years",[10118],{"type":26,"value":10119},"Why this takes years",{"type":21,"tag":22,"props":10121,"children":10122},{},[10123],{"type":26,"value":10124},"Migration is slow for reasons that have nothing to do with the algorithms:",{"type":21,"tag":3556,"props":10126,"children":10127},{},[10128,10138,10148,10158],{"type":21,"tag":3560,"props":10129,"children":10130},{},[10131,10136],{"type":21,"tag":71,"props":10132,"children":10133},{},[10134],{"type":26,"value":10135},"You don't know where your crypto is.",{"type":26,"value":10137}," It's in TLS terminators, JWT libraries, database TDE, backup encryption, VPN tunnels, code-signing pipelines, HSMs, IoT firmware, and vendor SDKs you don't control.",{"type":21,"tag":3560,"props":10139,"children":10140},{},[10141,10146],{"type":21,"tag":71,"props":10142,"children":10143},{},[10144],{"type":26,"value":10145},"Key and signature sizes change.",{"type":26,"value":10147}," ML-DSA public keys are roughly 1.3 KB versus 32 bytes for Ed25519, and SLH-DSA signatures sometimes run tens of kilobytes. Protocols with fixed-size fields, embedded devices with constrained flash, and certificate chains with MTU assumptions all break.",{"type":21,"tag":3560,"props":10149,"children":10150},{},[10151,10156],{"type":21,"tag":71,"props":10152,"children":10153},{},[10154],{"type":26,"value":10155},"Both endpoints must upgrade.",{"type":26,"value":10157}," You migrate at the speed of your slowest counterparty.",{"type":21,"tag":3560,"props":10159,"children":10160},{},[10161,10166],{"type":21,"tag":71,"props":10162,"children":10163},{},[10164],{"type":26,"value":10165},"Hardware has a decade-long tail.",{"type":26,"value":10167}," HSMs and embedded devices deployed today will still be running in 2035.",{"type":21,"tag":59,"props":10169,"children":10171},{"id":10170},"inventory-first-a-starting-point",[10172],{"type":26,"value":10173},"Inventory first: a starting point",{"type":21,"tag":22,"props":10175,"children":10176},{},[10177],{"type":26,"value":10178},"You cannot migrate an unknown. Start with automated discovery of what's negotiated in production, not what your architecture diagram claims.",{"type":21,"tag":312,"props":10180,"children":10184},{"className":10181,"code":10182,"language":10183,"meta":7,"style":7},"language-bash shiki shiki-themes github-dark","# What key exchange and signature algorithms is this endpoint using?\nopenssl s_client -connect example.com:443 -brief \u003C\u002Fdev\u002Fnull 2>&1 \\\n  | grep -Ei 'protocol|ciphersuite|group|peer signature'\n\n# Does the endpoint support hybrid post-quantum key exchange?\n# Requires OpenSSL 3.5+, which ships ML-KEM natively.\nopenssl s_client -connect example.com:443 \\\n  -groups X25519MLKEM768 \u003C\u002Fdev\u002Fnull 2>&1 | grep -i 'group\\|error'\n\n# Inventory certificate signature algorithms across an internal fleet\nwhile read -r host; do\n  alg=$(echo | openssl s_client -connect \"$host:443\" 2>\u002Fdev\u002Fnull \\\n    | openssl x509 -noout -text 2>\u002Fdev\u002Fnull \\\n    | awk '\u002FSignature Algorithm\u002F {print $3; exit}')\n  printf '%s\\t%s\\n' \"$host\" \"${alg:-UNREACHABLE}\"\ndone \u003C hosts.txt\n","bash",[10185],{"type":21,"tag":79,"props":10186,"children":10187},{"__ignoreMap":7},[10188,10196,10244,10267,10274,10282,10290,10313,10357,10364,10372,10405,10472,10511,10532,10582],{"type":21,"tag":322,"props":10189,"children":10190},{"class":324,"line":325},[10191],{"type":21,"tag":322,"props":10192,"children":10193},{"style":3888},[10194],{"type":26,"value":10195},"# What key exchange and signature algorithms is this endpoint using?\n",{"type":21,"tag":322,"props":10197,"children":10198},{"class":324,"line":335},[10199,10204,10209,10214,10219,10224,10229,10234,10239],{"type":21,"tag":322,"props":10200,"children":10201},{"style":4801},[10202],{"type":26,"value":10203},"openssl",{"type":21,"tag":322,"props":10205,"children":10206},{"style":359},[10207],{"type":26,"value":10208}," s_client",{"type":21,"tag":322,"props":10210,"children":10211},{"style":348},[10212],{"type":26,"value":10213}," -connect",{"type":21,"tag":322,"props":10215,"children":10216},{"style":359},[10217],{"type":26,"value":10218}," example.com:443",{"type":21,"tag":322,"props":10220,"children":10221},{"style":348},[10222],{"type":26,"value":10223}," -brief",{"type":21,"tag":322,"props":10225,"children":10226},{"style":3638},[10227],{"type":26,"value":10228}," \u003C",{"type":21,"tag":322,"props":10230,"children":10231},{"style":359},[10232],{"type":26,"value":10233},"\u002Fdev\u002Fnull",{"type":21,"tag":322,"props":10235,"children":10236},{"style":3638},[10237],{"type":26,"value":10238}," 2>&1",{"type":21,"tag":322,"props":10240,"children":10241},{"style":348},[10242],{"type":26,"value":10243}," \\\n",{"type":21,"tag":322,"props":10245,"children":10246},{"class":324,"line":344},[10247,10252,10257,10262],{"type":21,"tag":322,"props":10248,"children":10249},{"style":3638},[10250],{"type":26,"value":10251},"  |",{"type":21,"tag":322,"props":10253,"children":10254},{"style":4801},[10255],{"type":26,"value":10256}," grep",{"type":21,"tag":322,"props":10258,"children":10259},{"style":348},[10260],{"type":26,"value":10261}," -Ei",{"type":21,"tag":322,"props":10263,"children":10264},{"style":359},[10265],{"type":26,"value":10266}," 'protocol|ciphersuite|group|peer signature'\n",{"type":21,"tag":322,"props":10268,"children":10269},{"class":324,"line":407},[10270],{"type":21,"tag":322,"props":10271,"children":10272},{"emptyLinePlaceholder":3662},[10273],{"type":26,"value":3665},{"type":21,"tag":322,"props":10275,"children":10276},{"class":324,"line":431},[10277],{"type":21,"tag":322,"props":10278,"children":10279},{"style":3888},[10280],{"type":26,"value":10281},"# Does the endpoint support hybrid post-quantum key exchange?\n",{"type":21,"tag":322,"props":10283,"children":10284},{"class":324,"line":453},[10285],{"type":21,"tag":322,"props":10286,"children":10287},{"style":3888},[10288],{"type":26,"value":10289},"# Requires OpenSSL 3.5+, which ships ML-KEM natively.\n",{"type":21,"tag":322,"props":10291,"children":10292},{"class":324,"line":474},[10293,10297,10301,10305,10309],{"type":21,"tag":322,"props":10294,"children":10295},{"style":4801},[10296],{"type":26,"value":10203},{"type":21,"tag":322,"props":10298,"children":10299},{"style":359},[10300],{"type":26,"value":10208},{"type":21,"tag":322,"props":10302,"children":10303},{"style":348},[10304],{"type":26,"value":10213},{"type":21,"tag":322,"props":10306,"children":10307},{"style":359},[10308],{"type":26,"value":10218},{"type":21,"tag":322,"props":10310,"children":10311},{"style":348},[10312],{"type":26,"value":10243},{"type":21,"tag":322,"props":10314,"children":10315},{"class":324,"line":496},[10316,10321,10326,10330,10334,10338,10343,10347,10352],{"type":21,"tag":322,"props":10317,"children":10318},{"style":348},[10319],{"type":26,"value":10320},"  -groups",{"type":21,"tag":322,"props":10322,"children":10323},{"style":359},[10324],{"type":26,"value":10325}," X25519MLKEM768",{"type":21,"tag":322,"props":10327,"children":10328},{"style":3638},[10329],{"type":26,"value":10228},{"type":21,"tag":322,"props":10331,"children":10332},{"style":359},[10333],{"type":26,"value":10233},{"type":21,"tag":322,"props":10335,"children":10336},{"style":3638},[10337],{"type":26,"value":10238},{"type":21,"tag":322,"props":10339,"children":10340},{"style":3638},[10341],{"type":26,"value":10342}," |",{"type":21,"tag":322,"props":10344,"children":10345},{"style":4801},[10346],{"type":26,"value":10256},{"type":21,"tag":322,"props":10348,"children":10349},{"style":348},[10350],{"type":26,"value":10351}," -i",{"type":21,"tag":322,"props":10353,"children":10354},{"style":359},[10355],{"type":26,"value":10356}," 'group\\|error'\n",{"type":21,"tag":322,"props":10358,"children":10359},{"class":324,"line":16},[10360],{"type":21,"tag":322,"props":10361,"children":10362},{"emptyLinePlaceholder":3662},[10363],{"type":26,"value":3665},{"type":21,"tag":322,"props":10365,"children":10366},{"class":324,"line":3884},[10367],{"type":21,"tag":322,"props":10368,"children":10369},{"style":3888},[10370],{"type":26,"value":10371},"# Inventory certificate signature algorithms across an internal fleet\n",{"type":21,"tag":322,"props":10373,"children":10374},{"class":324,"line":3894},[10375,10380,10385,10390,10395,10400],{"type":21,"tag":322,"props":10376,"children":10377},{"style":3638},[10378],{"type":26,"value":10379},"while",{"type":21,"tag":322,"props":10381,"children":10382},{"style":348},[10383],{"type":26,"value":10384}," read",{"type":21,"tag":322,"props":10386,"children":10387},{"style":348},[10388],{"type":26,"value":10389}," -r",{"type":21,"tag":322,"props":10391,"children":10392},{"style":359},[10393],{"type":26,"value":10394}," host",{"type":21,"tag":322,"props":10396,"children":10397},{"style":329},[10398],{"type":26,"value":10399},"; ",{"type":21,"tag":322,"props":10401,"children":10402},{"style":3638},[10403],{"type":26,"value":10404},"do\n",{"type":21,"tag":322,"props":10406,"children":10407},{"class":324,"line":1034},[10408,10413,10417,10422,10427,10431,10436,10440,10444,10449,10454,10459,10464,10468],{"type":21,"tag":322,"props":10409,"children":10410},{"style":329},[10411],{"type":26,"value":10412},"  alg",{"type":21,"tag":322,"props":10414,"children":10415},{"style":3638},[10416],{"type":26,"value":3678},{"type":21,"tag":322,"props":10418,"children":10419},{"style":329},[10420],{"type":26,"value":10421},"$(",{"type":21,"tag":322,"props":10423,"children":10424},{"style":348},[10425],{"type":26,"value":10426},"echo",{"type":21,"tag":322,"props":10428,"children":10429},{"style":3638},[10430],{"type":26,"value":10342},{"type":21,"tag":322,"props":10432,"children":10433},{"style":4801},[10434],{"type":26,"value":10435}," openssl",{"type":21,"tag":322,"props":10437,"children":10438},{"style":359},[10439],{"type":26,"value":10208},{"type":21,"tag":322,"props":10441,"children":10442},{"style":348},[10443],{"type":26,"value":10213},{"type":21,"tag":322,"props":10445,"children":10446},{"style":359},[10447],{"type":26,"value":10448}," \"",{"type":21,"tag":322,"props":10450,"children":10451},{"style":329},[10452],{"type":26,"value":10453},"$host",{"type":21,"tag":322,"props":10455,"children":10456},{"style":359},[10457],{"type":26,"value":10458},":443\"",{"type":21,"tag":322,"props":10460,"children":10461},{"style":3638},[10462],{"type":26,"value":10463}," 2>",{"type":21,"tag":322,"props":10465,"children":10466},{"style":359},[10467],{"type":26,"value":10233},{"type":21,"tag":322,"props":10469,"children":10470},{"style":348},[10471],{"type":26,"value":10243},{"type":21,"tag":322,"props":10473,"children":10474},{"class":324,"line":3954},[10475,10480,10484,10489,10494,10499,10503,10507],{"type":21,"tag":322,"props":10476,"children":10477},{"style":3638},[10478],{"type":26,"value":10479},"    |",{"type":21,"tag":322,"props":10481,"children":10482},{"style":4801},[10483],{"type":26,"value":10435},{"type":21,"tag":322,"props":10485,"children":10486},{"style":359},[10487],{"type":26,"value":10488}," x509",{"type":21,"tag":322,"props":10490,"children":10491},{"style":348},[10492],{"type":26,"value":10493}," -noout",{"type":21,"tag":322,"props":10495,"children":10496},{"style":348},[10497],{"type":26,"value":10498}," -text",{"type":21,"tag":322,"props":10500,"children":10501},{"style":3638},[10502],{"type":26,"value":10463},{"type":21,"tag":322,"props":10504,"children":10505},{"style":359},[10506],{"type":26,"value":10233},{"type":21,"tag":322,"props":10508,"children":10509},{"style":348},[10510],{"type":26,"value":10243},{"type":21,"tag":322,"props":10512,"children":10513},{"class":324,"line":4002},[10514,10518,10523,10528],{"type":21,"tag":322,"props":10515,"children":10516},{"style":3638},[10517],{"type":26,"value":10479},{"type":21,"tag":322,"props":10519,"children":10520},{"style":4801},[10521],{"type":26,"value":10522}," awk",{"type":21,"tag":322,"props":10524,"children":10525},{"style":359},[10526],{"type":26,"value":10527}," '\u002FSignature Algorithm\u002F {print $3; exit}'",{"type":21,"tag":322,"props":10529,"children":10530},{"style":329},[10531],{"type":26,"value":3738},{"type":21,"tag":322,"props":10533,"children":10534},{"class":324,"line":4044},[10535,10540,10545,10549,10553,10557,10562,10567,10572,10577],{"type":21,"tag":322,"props":10536,"children":10537},{"style":348},[10538],{"type":26,"value":10539},"  printf",{"type":21,"tag":322,"props":10541,"children":10542},{"style":359},[10543],{"type":26,"value":10544}," '%s\\t%s\\n'",{"type":21,"tag":322,"props":10546,"children":10547},{"style":359},[10548],{"type":26,"value":10448},{"type":21,"tag":322,"props":10550,"children":10551},{"style":329},[10552],{"type":26,"value":10453},{"type":21,"tag":322,"props":10554,"children":10555},{"style":359},[10556],{"type":26,"value":3788},{"type":21,"tag":322,"props":10558,"children":10559},{"style":359},[10560],{"type":26,"value":10561}," \"${",{"type":21,"tag":322,"props":10563,"children":10564},{"style":329},[10565],{"type":26,"value":10566},"alg",{"type":21,"tag":322,"props":10568,"children":10569},{"style":3638},[10570],{"type":26,"value":10571},":-",{"type":21,"tag":322,"props":10573,"children":10574},{"style":329},[10575],{"type":26,"value":10576},"UNREACHABLE",{"type":21,"tag":322,"props":10578,"children":10579},{"style":359},[10580],{"type":26,"value":10581},"}\"\n",{"type":21,"tag":322,"props":10583,"children":10584},{"class":324,"line":5068},[10585,10590,10594],{"type":21,"tag":322,"props":10586,"children":10587},{"style":3638},[10588],{"type":26,"value":10589},"done",{"type":21,"tag":322,"props":10591,"children":10592},{"style":3638},[10593],{"type":26,"value":10228},{"type":21,"tag":322,"props":10595,"children":10596},{"style":329},[10597],{"type":26,"value":10598}," hosts.txt\n",{"type":21,"tag":22,"props":10600,"children":10601},{},[10602,10604,10609],{"type":26,"value":10603},"Feed the results into a register that records, per system: the algorithm, key size, protocol, certificate authority, library and version, the owner, and, most importantly, ",{"type":21,"tag":71,"props":10605,"children":10606},{},[10607],{"type":26,"value":10608},"how long the data it protects must stay confidential",{"type":26,"value":10610},". That last column is what turns an inventory into a prioritized migration order.",{"type":21,"tag":59,"props":10612,"children":10614},{"id":10613},"hybrid-deployment-is-the-low-risk-path",[10615],{"type":26,"value":10616},"Hybrid deployment is the low-risk path",{"type":21,"tag":22,"props":10618,"children":10619},{},[10620,10622,10627],{"type":26,"value":10621},"The consensus deployment pattern is hybrid: run a classical and a post-quantum key exchange together and combine both shared secrets, so the connection is safe unless ",{"type":21,"tag":51,"props":10623,"children":10624},{},[10625],{"type":26,"value":10626},"both",{"type":26,"value":10628}," are broken. This hedges against implementation bugs and undiscovered cryptanalysis in the newer lattice schemes.",{"type":21,"tag":22,"props":10630,"children":10631},{},[10632,10634,10639],{"type":26,"value":10633},"This is no longer experimental. ",{"type":21,"tag":71,"props":10635,"children":10636},{},[10637],{"type":26,"value":10638},"X25519MLKEM768",{"type":26,"value":10640}," is the standardized hybrid group, enabled by default in Chrome since version 131 and Firefox since 132, and OpenSSL 3.5 (April 2025) ships ML-KEM natively. Its default group list prefers the hybrid, meaning two OpenSSL 3.5 endpoints negotiate post-quantum key exchange with no configuration at all.",{"type":21,"tag":312,"props":10642,"children":10644},{"className":10181,"code":10643,"language":10183,"meta":7,"style":7},"# nginx with OpenSSL 3.5+: explicit hybrid preference, classical fallback\nssl_protocols TLSv1.3;\nssl_ecdh_curve X25519MLKEM768:X25519:P-256;\n",[10645],{"type":21,"tag":79,"props":10646,"children":10647},{"__ignoreMap":7},[10648,10656,10674],{"type":21,"tag":322,"props":10649,"children":10650},{"class":324,"line":325},[10651],{"type":21,"tag":322,"props":10652,"children":10653},{"style":3888},[10654],{"type":26,"value":10655},"# nginx with OpenSSL 3.5+: explicit hybrid preference, classical fallback\n",{"type":21,"tag":322,"props":10657,"children":10658},{"class":324,"line":335},[10659,10664,10669],{"type":21,"tag":322,"props":10660,"children":10661},{"style":4801},[10662],{"type":26,"value":10663},"ssl_protocols",{"type":21,"tag":322,"props":10665,"children":10666},{"style":359},[10667],{"type":26,"value":10668}," TLSv1.3",{"type":21,"tag":322,"props":10670,"children":10671},{"style":329},[10672],{"type":26,"value":10673},";\n",{"type":21,"tag":322,"props":10675,"children":10676},{"class":324,"line":344},[10677,10682,10687],{"type":21,"tag":322,"props":10678,"children":10679},{"style":4801},[10680],{"type":26,"value":10681},"ssl_ecdh_curve",{"type":21,"tag":322,"props":10683,"children":10684},{"style":359},[10685],{"type":26,"value":10686}," X25519MLKEM768:X25519:P-256",{"type":21,"tag":322,"props":10688,"children":10689},{"style":329},[10690],{"type":26,"value":10673},{"type":21,"tag":22,"props":10692,"children":10693},{},[10694],{"type":26,"value":10695},"Signatures are further behind. Public CAs are not yet issuing ML-DSA certificates at scale, so key exchange is where you get real HNDL protection today, and it's the cheapest win available.",{"type":21,"tag":59,"props":10697,"children":10699},{"id":10698},"what-to-do-this-quarter",[10700],{"type":26,"value":10701},"What to do this quarter",{"type":21,"tag":22,"props":10703,"children":10704},{},[10705],{"type":26,"value":10706},"A realistic 90-day slice:",{"type":21,"tag":7444,"props":10708,"children":10709},{},[10710,10720,10759,10769,10779,10789,10806],{"type":21,"tag":3560,"props":10711,"children":10712},{},[10713,10718],{"type":21,"tag":71,"props":10714,"children":10715},{},[10716],{"type":26,"value":10717},"Appoint an owner.",{"type":26,"value":10719}," Migrations without a named accountable person do not happen. This is a multi-year program, not a ticket.",{"type":21,"tag":3560,"props":10721,"children":10722},{},[10723,10728,10730,10736,10737,10743,10744,10750,10751,10757],{"type":21,"tag":71,"props":10724,"children":10725},{},[10726],{"type":26,"value":10727},"Run automated discovery",{"type":26,"value":10729}," across external endpoints, internal service mesh, and code repositories. Grep your codebases for ",{"type":21,"tag":79,"props":10731,"children":10733},{"className":10732},[],[10734],{"type":26,"value":10735},"RSA",{"type":26,"value":367},{"type":21,"tag":79,"props":10738,"children":10740},{"className":10739},[],[10741],{"type":26,"value":10742},"ECDSA",{"type":26,"value":367},{"type":21,"tag":79,"props":10745,"children":10747},{"className":10746},[],[10748],{"type":26,"value":10749},"secp256",{"type":26,"value":367},{"type":21,"tag":79,"props":10752,"children":10754},{"className":10753},[],[10755],{"type":26,"value":10756},"PKCS1",{"type":26,"value":10758},", and hardcoded algorithm identifiers.",{"type":21,"tag":3560,"props":10760,"children":10761},{},[10762,10767],{"type":21,"tag":71,"props":10763,"children":10764},{},[10765],{"type":26,"value":10766},"Classify by data lifetime.",{"type":26,"value":10768}," Sort systems by how long their data must stay secret. Anything over ten years is priority one.",{"type":21,"tag":3560,"props":10770,"children":10771},{},[10772,10777],{"type":21,"tag":71,"props":10773,"children":10774},{},[10775],{"type":26,"value":10776},"Turn on hybrid TLS where it's free.",{"type":26,"value":10778}," Upgrade to OpenSSL 3.5+ or a TLS terminator that supports X25519MLKEM768. For many organizations this is a load-balancer config change covering the majority of traffic.",{"type":21,"tag":3560,"props":10780,"children":10781},{},[10782,10787],{"type":21,"tag":71,"props":10783,"children":10784},{},[10785],{"type":26,"value":10786},"Send the vendor questionnaire.",{"type":26,"value":10788}," Ask every crypto-touching vendor for their FIPS 203\u002F204\u002F205 roadmap with dates. Slow answers are your long pole. You need them on record now.",{"type":21,"tag":3560,"props":10790,"children":10791},{},[10792,10797,10799,10804],{"type":21,"tag":71,"props":10793,"children":10794},{},[10795],{"type":26,"value":10796},"Fix crypto-agility in new code.",{"type":26,"value":10798}," Stop hardcoding algorithms. Route every operation through an abstraction that takes the algorithm from configuration, so the ",{"type":21,"tag":51,"props":10800,"children":10801},{},[10802],{"type":26,"value":10803},"next",{"type":26,"value":10805}," transition (HQC, FN-DSA, or whatever follows) is a config change rather than a rewrite. This is the single highest-leverage engineering change, because the 2035 deadline is not the last one.",{"type":21,"tag":3560,"props":10807,"children":10808},{},[10809,10814],{"type":21,"tag":71,"props":10810,"children":10811},{},[10812],{"type":26,"value":10813},"Move code signing early.",{"type":26,"value":10815}," It has the earliest federal deadline, the longest verification tail, and signatures that must remain verifiable for decades.",{"type":21,"tag":59,"props":10817,"children":10819},{"id":10818},"the-bottom-line",[10820],{"type":26,"value":10821},"The bottom line",{"type":21,"tag":22,"props":10823,"children":10824},{},[10825,10827,10832],{"type":26,"value":10826},"The deadlines are published, they're closer than the hardware timeline suggests, and they were set by people who assume migration takes ten years. Two dates are worth memorizing: ",{"type":21,"tag":71,"props":10828,"children":10829},{},[10830],{"type":26,"value":10831},"after 2030, classical asymmetric crypto is deprecated, and after 2035, it's disallowed",{"type":26,"value":10833},", with federal national-security systems running years ahead of that.",{"type":21,"tag":22,"props":10835,"children":10836},{},[10837],{"type":26,"value":10838},"The work that pays off regardless of when quantum hardware arrives is the boring part: knowing where your cryptography lives and being able to change it without a rewrite. Everything else is a config flag.",{"type":21,"tag":22,"props":10840,"children":10841},{},[10842,10847,10848,10853,10854,10859,10860,10865,10866],{"type":21,"tag":71,"props":10843,"children":10844},{},[10845],{"type":26,"value":10846},"Further reading:",{"type":26,"value":1901},{"type":21,"tag":29,"props":10849,"children":10850},{"href":9817},[10851],{"type":26,"value":10852},"What PQC is",{"type":26,"value":7549},{"type":21,"tag":29,"props":10855,"children":10856},{"href":1274},[10857],{"type":26,"value":10858},"Reading policy documents as technical assessments",{"type":26,"value":7549},{"type":21,"tag":29,"props":10861,"children":10862},{"href":4400},[10863],{"type":26,"value":10864},"Quantum computing glossary",{"type":26,"value":7549},{"type":21,"tag":29,"props":10867,"children":10869},{"href":10868},"\u002Fnews",[10870],{"type":26,"value":10871},"Latest developments",{"type":21,"tag":703,"props":10873,"children":10874},{},[10875],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":10877},[10878,10879,10880,10881,10882,10883,10884,10885],{"id":9825,"depth":335,"text":9828},{"id":9970,"depth":335,"text":9973},{"id":10048,"depth":335,"text":10051},{"id":10116,"depth":335,"text":10119},{"id":10170,"depth":335,"text":10173},{"id":10613,"depth":335,"text":10616},{"id":10698,"depth":335,"text":10701},{"id":10818,"depth":335,"text":10821},"content:blog:post-quantum-migration-deadlines.md","blog\u002Fpost-quantum-migration-deadlines.md","blog\u002Fpost-quantum-migration-deadlines",{"_path":10890,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":10891,"description":10892,"date":3500,"author":11,"tags":10893,"readingTime":16,"body":10894,"_type":715,"_id":11460,"_source":717,"_file":11461,"_stem":11462,"_extension":720},"\u002Fblog\u002Fq-day-breaking-rsa-2048","Q-Day and What Breaking RSA-2048 Requires","Q-Day is one of the most abused terms in quantum computing. Here's a precise definition, the published resource estimates for factoring RSA-2048 with Shor's algorithm, and why those estimates disagree by orders of magnitude.",[9785,9786,9787],{"type":18,"children":10895,"toc":11450},[10896,10901,10906,10912,10936,10948,10953,10959,10977,10989,10995,11015,11054,11083,11089,11123,11150,11176,11202,11207,11260,11270,11276,11300,11312,11318,11323,11335,11341,11346,11372,11384,11390,11395,11407,11419],{"type":21,"tag":22,"props":10897,"children":10898},{},[10899],{"type":26,"value":10900},"\"Q-Day\" appears in vendor decks, government strategy documents, and newspaper headlines, usually with a countdown clock attached. It is rarely defined. That vagueness is doing a lot of work, because the term is used to mean at least two different things, and the gap between them is measured in years and billions of dollars of engineering.",{"type":21,"tag":22,"props":10902,"children":10903},{},[10904],{"type":26,"value":10905},"This post does the part that most coverage skips: what a machine capable of breaking RSA-2048 would have to be, according to published estimates, and why those estimates have ranged from twenty million qubits to under one million within six years.",{"type":21,"tag":59,"props":10907,"children":10909},{"id":10908},"defining-the-term-precisely",[10910],{"type":26,"value":10911},"Defining the term precisely",{"type":21,"tag":22,"props":10913,"children":10914},{},[10915,10921,10923,10928,10930,10934],{"type":21,"tag":29,"props":10916,"children":10918},{"href":10917},"\u002Fglossary\u002Fq-day",[10919],{"type":26,"value":10920},"Q-Day",{"type":26,"value":10922}," is most usefully defined as ",{"type":21,"tag":71,"props":10924,"children":10925},{},[10926],{"type":26,"value":10927},"the day a cryptographically relevant quantum computer (CRQC) first exists",{"type":26,"value":10929},": a machine that runs ",{"type":21,"tag":29,"props":10931,"children":10932},{"href":10066},[10933],{"type":26,"value":10069},{"type":26,"value":10935}," against real-world key sizes within an operationally useful time.",{"type":21,"tag":22,"props":10937,"children":10938},{},[10939,10941,10946],{"type":26,"value":10940},"The looseness creeps in around \"real-world key sizes.\" Some writers use Q-Day to mean specifically ",{"type":21,"tag":51,"props":10942,"children":10943},{},[10944],{"type":26,"value":10945},"RSA-2048 is broken",{"type":26,"value":10947},". Others use it for any CRQC, which would include a machine that breaks NIST P-256 elliptic-curve keys, a meaningfully easier target, since ECC uses far smaller keys for equivalent classical security, and the elliptic-curve variant of Shor's needs a smaller register. A third group uses it loosely for \"quantum computers become scary,\" which is not a technical claim at all.",{"type":21,"tag":22,"props":10949,"children":10950},{},[10951],{"type":26,"value":10952},"Precision matters because these events do not happen on the same day. If P-256 falls before RSA-2048, the practical consequences differ sharply: ECDSA underpins TLS certificates, SSH, code signing, and most cryptocurrency, while RSA-2048 still guards a large installed base of older infrastructure. A plan built around one date risks being badly calibrated for the other.",{"type":21,"tag":59,"props":10954,"children":10956},{"id":10955},"why-the-date-matters-less-than-youd-think",[10957],{"type":26,"value":10958},"Why the date matters less than you'd think",{"type":21,"tag":22,"props":10960,"children":10961},{},[10962,10964,10968,10970,10975],{"type":26,"value":10963},"The instinct on hearing \"Q-Day\" is to ask when. That question is less decision-relevant than it feels, because of ",{"type":21,"tag":71,"props":10965,"children":10966},{},[10967],{"type":26,"value":10061},{"type":26,"value":10969},": an adversary records encrypted traffic today and stores it until a CRQC exists. Any data whose confidentiality must outlive Q-Day is exposed ",{"type":21,"tag":51,"props":10971,"children":10972},{},[10973],{"type":26,"value":10974},"now",{"type":26,"value":10976},", regardless of when Q-Day lands.",{"type":21,"tag":22,"props":10978,"children":10979},{},[10980,10982,10987],{"type":26,"value":10981},"So the question that drives decisions is not \"when is Q-Day\" but \"how long must this data stay secret, and how long will my migration take.\" Regulators have already answered on organizations' behalf with published deprecation dates. We cover those in detail in ",{"type":21,"tag":29,"props":10983,"children":10984},{"href":2343},[10985],{"type":26,"value":10986},"the post-quantum migration clock",{"type":26,"value":10988},", and they are deliberately independent of any hardware forecast.",{"type":21,"tag":59,"props":10990,"children":10992},{"id":10991},"the-core-question-what-would-the-machine-look-like",[10993],{"type":26,"value":10994},"The core question: what would the machine look like?",{"type":21,"tag":22,"props":10996,"children":10997},{},[10998,11000,11005,11007,11013],{"type":26,"value":10999},"Shor's algorithm, ",{"type":21,"tag":29,"props":11001,"children":11002},{"href":9045},[11003],{"type":26,"value":11004},"introduced in 1994",{"type":26,"value":11006},", reduces factoring to period-finding, which it solves using ",{"type":21,"tag":29,"props":11008,"children":11010},{"href":11009},"\u002Fglossary\u002Fphase-estimation",[11011],{"type":26,"value":11012},"quantum phase estimation",{"type":26,"value":11014}," over a modular exponentiation circuit. The algorithmic cost is polynomial. The engineering cost is where the difficulty lives, and it splits into two different numbers.",{"type":21,"tag":22,"props":11016,"children":11017},{},[11018,11023,11025,11031,11033,11037,11039,11044,11046,11052],{"type":21,"tag":71,"props":11019,"children":11020},{},[11021],{"type":26,"value":11022},"Logical qubits",{"type":26,"value":11024}," are the idealized, error-free qubits the algorithm is written against. Gidney and Ekerå's 2019 construction expresses this as ",{"type":21,"tag":79,"props":11026,"children":11028},{"className":11027},[],[11029],{"type":26,"value":11030},"3n + 0.002·n·lg n",{"type":26,"value":11032}," logical qubits for an ",{"type":21,"tag":51,"props":11034,"children":11035},{},[11036],{"type":26,"value":5266},{"type":26,"value":11038},"-bit modulus. For RSA-2048 that works out to roughly ",{"type":21,"tag":71,"props":11040,"children":11041},{},[11042],{"type":26,"value":11043},"6,200 logical qubits",{"type":26,"value":11045},", alongside a Toffoli gate count on the order of ",{"type":21,"tag":79,"props":11047,"children":11049},{"className":11048},[],[11050],{"type":26,"value":11051},"0.3n³",{"type":26,"value":11053},", billions of gates.",{"type":21,"tag":22,"props":11055,"children":11056},{},[11057,11062,11064,11068,11070,11074,11076,11081],{"type":21,"tag":71,"props":11058,"children":11059},{},[11060],{"type":26,"value":11061},"Physical qubits",{"type":26,"value":11063}," are what you build. The ratio between them is the overhead imposed by ",{"type":21,"tag":29,"props":11065,"children":11066},{"href":1526},[11067],{"type":26,"value":9402},{"type":26,"value":11069},", and it is brutal. Under the surface code, each ",{"type":21,"tag":29,"props":11071,"children":11072},{"href":1983},[11073],{"type":26,"value":9390},{"type":26,"value":11075}," is encoded across a patch of physical qubits whose size grows with the ",{"type":21,"tag":51,"props":11077,"children":11078},{},[11079],{"type":26,"value":11080},"code distance",{"type":26,"value":11082},", which must be chosen large enough that the total error across billions of gates stays below one. Getting from ~6,200 logical qubits to a working machine is where the millions come from.",{"type":21,"tag":59,"props":11084,"children":11086},{"id":11085},"the-published-estimates-and-why-they-moved",[11087],{"type":26,"value":11088},"The published estimates, and why they moved",{"type":21,"tag":22,"props":11090,"children":11091},{},[11092,11094,11099,11101,11106,11107,11114,11116,11121],{"type":26,"value":11093},"The most-cited figure comes from ",{"type":21,"tag":71,"props":11095,"children":11096},{},[11097],{"type":26,"value":11098},"Gidney and Ekerå (2019)",{"type":26,"value":11100},": factoring a 2048-bit RSA integer in ",{"type":21,"tag":71,"props":11102,"children":11103},{},[11104],{"type":26,"value":11105},"8 hours using 20 million noisy physical qubits",{"type":26,"value":9988},{"type":21,"tag":29,"props":11108,"children":11111},{"href":11109,"rel":11110},"https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.09749",[44],[11112],{"type":26,"value":11113},"arXiv:1905.09749",{"type":26,"value":11115},", later published in ",{"type":21,"tag":51,"props":11117,"children":11118},{},[11119],{"type":26,"value":11120},"Quantum",{"type":26,"value":11122},"). That was already a hundredfold reduction in spacetime volume over prior proposals.",{"type":21,"tag":22,"props":11124,"children":11125},{},[11126,11128,11133,11135,11140,11141,11148],{"type":26,"value":11127},"In May 2025, ",{"type":21,"tag":71,"props":11129,"children":11130},{},[11131],{"type":26,"value":11132},"Gidney",{"type":26,"value":11134}," published a revised analysis: ",{"type":21,"tag":71,"props":11136,"children":11137},{},[11138],{"type":26,"value":11139},"fewer than one million noisy physical qubits, running for under a week",{"type":26,"value":9988},{"type":21,"tag":29,"props":11142,"children":11145},{"href":11143,"rel":11144},"https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.15917",[44],[11146],{"type":26,"value":11147},"arXiv:2505.15917",{"type":26,"value":11149},"). That is roughly a 20× reduction in qubit count against his own earlier number, traded for a longer runtime.",{"type":21,"tag":22,"props":11151,"children":11152},{},[11153,11155,11160,11162,11167,11169,11174],{"type":26,"value":11154},"Both papers assume broadly the same hardware model: a planar square grid of superconducting qubits with nearest-neighbour connectivity, a ",{"type":21,"tag":71,"props":11156,"children":11157},{},[11158],{"type":26,"value":11159},"0.1% gate error rate",{"type":26,"value":11161},", a ",{"type":21,"tag":71,"props":11163,"children":11164},{},[11165],{"type":26,"value":11166},"1 microsecond",{"type":26,"value":11168}," surface code cycle, and a ",{"type":21,"tag":71,"props":11170,"children":11171},{},[11172],{"type":26,"value":11173},"10 microsecond",{"type":26,"value":11175}," control-system reaction time.",{"type":21,"tag":22,"props":11177,"children":11178},{},[11179,11181,11186,11188,11193,11195,11200],{"type":26,"value":11180},"That is the key insight about the discrepancy: ",{"type":21,"tag":71,"props":11182,"children":11183},{},[11184],{"type":26,"value":11185},"the hardware assumptions barely changed. The algorithms and codes did.",{"type":26,"value":11187}," The 2025 reduction comes from three specific advances: approximate residue arithmetic (fewer operations), ",{"type":21,"tag":51,"props":11189,"children":11190},{},[11191],{"type":26,"value":11192},"yoked surface codes",{"type":26,"value":11194}," (roughly tripling storage density for idle logical qubits), and ",{"type":21,"tag":51,"props":11196,"children":11197},{},[11198],{"type":26,"value":11199},"magic state cultivation",{"type":26,"value":11201},", which makes producing the high-fidelity non-Clifford resource states that dominate the cost far cheaper than traditional distillation.",{"type":21,"tag":22,"props":11203,"children":11204},{},[11205],{"type":26,"value":11206},"This is why estimates in the wild vary so wildly. Any published number is a function of at least five assumptions:",{"type":21,"tag":3556,"props":11208,"children":11209},{},[11210,11220,11230,11240,11250],{"type":21,"tag":3560,"props":11211,"children":11212},{},[11213,11218],{"type":21,"tag":71,"props":11214,"children":11215},{},[11216],{"type":26,"value":11217},"Physical error rate.",{"type":26,"value":11219}," The required code distance depends on how far below threshold the hardware sits. A 10× better physical error rate cuts the overhead dramatically.",{"type":21,"tag":3560,"props":11221,"children":11222},{},[11223,11228],{"type":21,"tag":71,"props":11224,"children":11225},{},[11226],{"type":26,"value":11227},"Code choice.",{"type":26,"value":11229}," Surface codes are the conservative baseline. Higher-rate qLDPC codes are likely to reduce overhead substantially, at the cost of harder connectivity requirements.",{"type":21,"tag":3560,"props":11231,"children":11232},{},[11233,11238],{"type":21,"tag":71,"props":11234,"children":11235},{},[11236],{"type":26,"value":11237},"Cycle and reaction time.",{"type":26,"value":11239}," These set wall-clock runtime, and runtime feeds back into how much error correction you need.",{"type":21,"tag":3560,"props":11241,"children":11242},{},[11243,11248],{"type":21,"tag":71,"props":11244,"children":11245},{},[11246],{"type":26,"value":11247},"Magic state strategy.",{"type":26,"value":11249}," Distillation factories historically consumed the majority of the machine's footprint. Cultivation changes that budget.",{"type":21,"tag":3560,"props":11251,"children":11252},{},[11253,11258],{"type":21,"tag":71,"props":11254,"children":11255},{},[11256],{"type":26,"value":11257},"Space–time tradeoff.",{"type":26,"value":11259}," You nearly always buy fewer qubits with more hours, which is exactly what happened between the 2019 and 2025 papers. Quoting a qubit count without a runtime is meaningless.",{"type":21,"tag":22,"props":11261,"children":11262},{},[11263,11268],{"type":21,"tag":71,"props":11264,"children":11265},{},[11266],{"type":26,"value":11267},"The direction of travel is the important dynamic.",{"type":26,"value":11269}," Over three decades, published estimates have moved consistently downward, driven by compilation and error-correction improvements rather than by hardware surprises. A migration plan that assumes today's estimate is the floor is planning against a number that has never yet stopped falling.",{"type":21,"tag":59,"props":11271,"children":11273},{"id":11272},"where-hardware-stands",[11274],{"type":26,"value":11275},"Where hardware stands",{"type":21,"tag":22,"props":11277,"children":11278},{},[11279,11281,11286,11288,11292,11294,11298],{"type":26,"value":11280},"Current devices carry roughly 100 to 1,000 physical qubits. Google's Willow processor, used in the 2024 ",{"type":21,"tag":29,"props":11282,"children":11283},{"href":4184},[11284],{"type":26,"value":11285},"below-threshold demonstration",{"type":26,"value":11287},", has 105. Even against the optimistic 2025 estimate, that is three orders of magnitude short in qubit count, before accounting for the ",{"type":21,"tag":29,"props":11289,"children":11290},{"href":9450},[11291],{"type":26,"value":9453},{"type":26,"value":11293},", connectivity, and control requirements those papers assume. Our ",{"type":21,"tag":29,"props":11295,"children":11296},{"href":806},[11297],{"type":26,"value":1706},{"type":26,"value":11299}," tracks what is available.",{"type":21,"tag":22,"props":11301,"children":11302},{},[11303,11305,11310],{"type":26,"value":11304},"But the 2024 result matters more than the raw gap suggests. Google ran surface codes at distances 3, 5, and 7 and showed each increase roughly ",{"type":21,"tag":51,"props":11306,"children":11307},{},[11308],{"type":26,"value":11309},"halved",{"type":26,"value":11311}," the logical error rate: the logical qubit outperformed its best constituent physical qubit. Every fault-tolerance roadmap assumes that adding qubits makes the encoded qubit better rather than worse. Until that experiment, it was an assumption. Crossing the threshold converted the remaining problem from an open physics question into a scaling and manufacturing one. Scaling problems are still extremely hard and sometimes take decades. They are a different category of hard.",{"type":21,"tag":59,"props":11313,"children":11315},{"id":11314},"machine-checking-the-numbers",[11316],{"type":26,"value":11317},"Machine-checking the numbers",{"type":21,"tag":22,"props":11319,"children":11320},{},[11321],{"type":26,"value":11322},"A subtlety that rarely surfaces: these resource estimates are long, detailed, hand-written analyses, and migration budgets running into the billions are set on the basis of them. An arithmetic slip or an unstated assumption in a compilation argument propagates straight into policy.",{"type":21,"tag":22,"props":11324,"children":11325},{},[11326,11328,11333],{"type":26,"value":11327},"That is why the 2026 ",{"type":21,"tag":29,"props":11329,"children":11330},{"href":10097},[11331],{"type":26,"value":11332},"Lean formalization of Shor's algorithm",{"type":26,"value":11334}," is worth attention beyond its novelty. It formalizes order finding and the reversible circuits for modular and elliptic-curve arithmetic in a proof assistant, and machine-checks the logical resource estimates for both RSA-2048 and P-256. It does not settle the physical-qubit question (that still depends on the hardware assumptions above), but it puts the logical layer on an auditable footing. When a number drives a budget, \"a machine verified this derivation\" is a meaningfully stronger claim than \"several experts read the appendix.\"",{"type":21,"tag":59,"props":11336,"children":11338},{"id":11337},"the-vendor-perspective-clearly-labelled",[11339],{"type":26,"value":11340},"The vendor perspective, clearly labelled",{"type":21,"tag":22,"props":11342,"children":11343},{},[11344],{"type":26,"value":11345},"Quantum hardware companies publish extensively on Q-Day, and their framing deserves an explicit caveat: they sell the machines whose urgency they are describing.",{"type":21,"tag":22,"props":11347,"children":11348},{},[11349,11351,11356,11358,11363,11365,11370],{"type":26,"value":11350},"IonQ's December 2024 post ",{"type":21,"tag":51,"props":11352,"children":11353},{},[11354],{"type":26,"value":11355},"Q-Day and the Impact of Breaking RSA2048",{"type":26,"value":11357},", by SVP of Product and Applications Ariel Braunstein, is a reasonable example. ",{"type":21,"tag":71,"props":11359,"children":11360},{},[11361],{"type":26,"value":11362},"IonQ argues",{"type":26,"value":11364}," that harvest-now-decrypt-later is already underway, describing nation states collecting encrypted data in anticipation of future decryption capability, and that the unpredictability of algorithmic breakthroughs is itself a reason not to wait. To its credit, the post declines to name a year, presenting a spread of estimates across research organizations instead, and ",{"type":21,"tag":71,"props":11366,"children":11367},{},[11368],{"type":26,"value":11369},"IonQ's recommendation",{"type":26,"value":11371}," (that organizations begin exploring quantum-resistant algorithms now, while acknowledging urgency varies with the timeline one assumes) is close to the mainstream security-community position.",{"type":21,"tag":22,"props":11373,"children":11374},{},[11375,11377,11382],{"type":26,"value":11376},"Note also what it does not contain. ",{"type":21,"tag":71,"props":11378,"children":11379},{},[11380],{"type":26,"value":11381},"The IonQ post cites no qubit count or resource estimate for RSA-2048 at all.",{"type":26,"value":11383}," Its one concrete figure is an unrelated materials-simulation algorithm reduced from 1.5 trillion gate operations to 410,000. That is a genuine result about a different workload, and it is not evidence about factoring. Treat vendor content as a directionally useful signal about industry expectations, not as a neutral source for the resource question.",{"type":21,"tag":59,"props":11385,"children":11387},{"id":11386},"how-to-reason-about-it-yourself",[11388],{"type":26,"value":11389},"How to reason about it yourself",{"type":21,"tag":22,"props":11391,"children":11392},{},[11393],{"type":26,"value":11394},"The framing that survives contact with the uncertainty is Mosca's inequality: if your data's required confidentiality lifetime plus your migration time exceeds the time until a CRQC exists, you are already late. Only the third term is unknown, and the first two are usually large enough that the third barely changes the answer.",{"type":21,"tag":22,"props":11396,"children":11397},{},[11398,11400,11405],{"type":26,"value":11399},"So the honest conclusion is an uncomfortable one: ",{"type":21,"tag":71,"props":11401,"children":11402},{},[11403],{"type":26,"value":11404},"nobody credible gives you a date, and anyone who does is telling you about their business model or their priors, not about physics.",{"type":26,"value":11406}," What we say precisely is what the machine would need to be, that the requirement has been revised downward repeatedly, and that the error-correction threshold has been crossed.",{"type":21,"tag":22,"props":11408,"children":11409},{},[11410,11412,11417],{"type":26,"value":11411},"The responsible posture is to migrate on a schedule that does not depend on knowing the date. Start with ",{"type":21,"tag":29,"props":11413,"children":11414},{"href":9817},[11415],{"type":26,"value":11416},"what PQC is",{"type":26,"value":11418},", then work the calendar. If your plan breaks when the estimate drops another 10×, it was never a plan.",{"type":21,"tag":22,"props":11420,"children":11421},{},[11422,11426,11427,11432,11433,11438,11439,11445,11446],{"type":21,"tag":71,"props":11423,"children":11424},{},[11425],{"type":26,"value":10846},{"type":26,"value":1901},{"type":21,"tag":29,"props":11428,"children":11429},{"href":2343},[11430],{"type":26,"value":11431},"Post-quantum migration deadlines",{"type":26,"value":7549},{"type":21,"tag":29,"props":11434,"children":11435},{"href":854},[11436],{"type":26,"value":11437},"Quantum error correction explained",{"type":26,"value":7549},{"type":21,"tag":29,"props":11440,"children":11442},{"href":11441},"\u002Fresearch",[11443],{"type":26,"value":11444},"Landmark research papers",{"type":26,"value":7549},{"type":21,"tag":29,"props":11447,"children":11448},{"href":10868},[11449],{"type":26,"value":10871},{"title":7,"searchDepth":335,"depth":335,"links":11451},[11452,11453,11454,11455,11456,11457,11458,11459],{"id":10908,"depth":335,"text":10911},{"id":10955,"depth":335,"text":10958},{"id":10991,"depth":335,"text":10994},{"id":11085,"depth":335,"text":11088},{"id":11272,"depth":335,"text":11275},{"id":11314,"depth":335,"text":11317},{"id":11337,"depth":335,"text":11340},{"id":11386,"depth":335,"text":11389},"content:blog:q-day-breaking-rsa-2048.md","blog\u002Fq-day-breaking-rsa-2048.md","blog\u002Fq-day-breaking-rsa-2048",{"_path":1109,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":11464,"description":11465,"date":3500,"author":11,"tags":11466,"readingTime":16,"body":11467,"_type":715,"_id":11959,"_source":717,"_file":11960,"_stem":11961,"_extension":720},"The Real-Time Decoding Bottleneck: Quantum Error Correction's Classical Problem","Syndrome measurement tells you something broke, not what broke. Working that out is decoding, and if your classical decoder can't keep pace with the syndrome stream, the whole error-correction scheme collapses.",[729,730,3502],{"type":18,"children":11468,"toc":11949},[11469,11486,11498,11504,11516,11529,11541,11553,11559,11564,11569,11574,11586,11598,11610,11622,11628,11633,11643,11653,11663,11669,11674,11679,11691,11710,11715,11720,11726,11738,11743,11774,11786,11792,11797,11829,11834,11840,11852,11857,11876,11882,11887,11892,11903],{"type":21,"tag":22,"props":11470,"children":11471},{},[11472,11474,11478,11480,11484],{"type":26,"value":11473},"Most explanations of ",{"type":21,"tag":29,"props":11475,"children":11476},{"href":1526},[11477],{"type":26,"value":9402},{"type":26,"value":11479}," end at the same place. You encode a logical qubit across many physical ones, you run parity checks with ",{"type":21,"tag":29,"props":11481,"children":11482},{"href":9414},[11483],{"type":26,"value":9417},{"type":26,"value":11485},", you detect errors without collapsing the state, and then the article stops, having implied that detection is correction.",{"type":21,"tag":22,"props":11487,"children":11488},{},[11489,11491,11496],{"type":26,"value":11490},"It isn't. In our ",{"type":21,"tag":29,"props":11492,"children":11493},{"href":751},[11494],{"type":26,"value":11495},"companion post on logical qubits and fault tolerance",{"type":26,"value":11497},", we listed real-time decoding as one of four things a fault-tolerant machine needs, and then moved on. This post picks up that thread, because it is the requirement least likely to be explained anywhere else, and it has a property the others don't: it is not fundamentally a quantum problem at all. It is a classical computing problem sitting inside the quantum computer, and it is currently one of the harder ones.",{"type":21,"tag":59,"props":11499,"children":11501},{"id":11500},"what-syndrome-measurement-gives-you",[11502],{"type":26,"value":11503},"What syndrome measurement gives you",{"type":21,"tag":22,"props":11505,"children":11506},{},[11507,11509,11514],{"type":26,"value":11508},"The ",{"type":21,"tag":29,"props":11510,"children":11511},{"href":854},[11512],{"type":26,"value":11513},"QEC basics post",{"type":26,"value":11515}," covers the mechanics of parity checks, so we won't re-derive them. The relevant point is what comes out.",{"type":21,"tag":22,"props":11517,"children":11518},{},[11519,11521,11527],{"type":26,"value":11520},"A round of syndrome extraction hands you a list of bits. Each bit says whether one particular parity check agreed with its previous value or flipped. A flipped check means an error touched at least one of the qubits it monitors. That's it. You do not learn which qubit. You do not learn whether it was an X, Z, or Y error. See the ",{"type":21,"tag":29,"props":11522,"children":11524},{"href":11523},"\u002Fglossary\u002Fpauli-operators",[11525],{"type":26,"value":11526},"Pauli operator",{"type":26,"value":11528}," breakdown for why those are the only cases that matter. You learn only that the parity structure has been violated somewhere in the neighbourhood of that check.",{"type":21,"tag":22,"props":11530,"children":11531},{},[11532,11534,11539],{"type":26,"value":11533},"Worse, the ",{"type":21,"tag":29,"props":11535,"children":11536},{"href":7857},[11537],{"type":26,"value":11538},"measurement",{"type":26,"value":11540}," circuits are themselves noisy, so some flipped checks correspond to no data error at all. The check itself misfired. And errors that happen to straddle two checks flip both, while an error in the middle of a chain of errors sometimes flips nothing, because two violations cancel.",{"type":21,"tag":22,"props":11542,"children":11543},{},[11544,11546,11551],{"type":26,"value":11545},"So the syndrome is not a diagnosis. It is a constraint. The job is to find the most probable physical error consistent with that constraint, then apply (or, more often, simply record) the correction. That inference step is ",{"type":21,"tag":71,"props":11547,"children":11548},{},[11549],{"type":26,"value":11550},"decoding",{"type":26,"value":11552},", and it is a genuine combinatorial optimisation problem.",{"type":21,"tag":59,"props":11554,"children":11556},{"id":11555},"the-part-that-makes-it-urgent",[11557],{"type":26,"value":11558},"The part that makes it urgent",{"type":21,"tag":22,"props":11560,"children":11561},{},[11562],{"type":26,"value":11563},"Here is where it stops being an interesting algorithms question and becomes an engineering crisis.",{"type":21,"tag":22,"props":11565,"children":11566},{},[11567],{"type":26,"value":11568},"Syndrome extraction is not a one-off. It runs continuously, in rounds, for the entire duration of the computation. On superconducting hardware a round takes roughly a microsecond. Google's Willow experiments ran a cycle time of about 1.1 µs. Every one of those cycles emits a fresh syndrome frame from every logical qubit on the chip. For a machine of any interesting size, the syndrome data rate runs into millions of measurements per second.",{"type":21,"tag":22,"props":11570,"children":11571},{},[11572],{"type":26,"value":11573},"Your decoder has to consume that stream at least as fast as it arrives. Not \"reasonably fast.\" At least as fast, sustained, forever.",{"type":21,"tag":22,"props":11575,"children":11576},{},[11577,11579,11584],{"type":26,"value":11578},"If it doesn't, you get the ",{"type":21,"tag":71,"props":11580,"children":11581},{},[11582],{"type":26,"value":11583},"backlog problem",{"type":26,"value":11585},", and the backlog problem does not degrade gracefully. Suppose your decoder processes each round 10% slower than rounds are produced. Undecoded frames pile up. The queue grows linearly in wall-clock time, which sounds survivable until you remember why you needed the decode result in the first place.",{"type":21,"tag":22,"props":11587,"children":11588},{},[11589,11591,11596],{"type":26,"value":11590},"Certain operations in a fault-tolerant computation are ",{"type":21,"tag":51,"props":11592,"children":11593},{},[11594],{"type":26,"value":11595},"conditional",{"type":26,"value":11597},": what gate you apply next depends on the decoded outcome of measurements already taken. The canonical case is a non-Clifford gate implemented via magic state injection, where a correction is applied or not depending on a measurement result that must first be decoded. At each such branch point, the machine must stop and wait for the decoder to catch up to the present moment. But while it waits, more syndrome rounds are being produced. The qubits don't stop decohering only because the classical side is busy. So the wait itself extends the backlog, which lengthens the next wait, which extends the backlog further.",{"type":21,"tag":22,"props":11599,"children":11600},{},[11601,11603,11608],{"type":26,"value":11602},"The result is an ",{"type":21,"tag":71,"props":11604,"children":11605},{},[11606],{"type":26,"value":11607},"exponential slowdown",{"type":26,"value":11609},". Each conditional operation takes longer than the last by a compounding factor, and the computation grinds to a halt long before it finishes. A decoder that is 10% too slow doesn't cost you 10%. It costs you the entire computation. The requirement is a hard throughput threshold, not a performance preference.",{"type":21,"tag":22,"props":11611,"children":11612},{},[11613,11615,11620],{"type":26,"value":11614},"This is why the word \"real-time\" is load-bearing. Offline decoding (recording the syndrome stream and analysing it afterwards) is perfectly fine for a ",{"type":21,"tag":51,"props":11616,"children":11617},{},[11618],{"type":26,"value":11619},"memory",{"type":26,"value":11621}," experiment, where you only want to know whether the stored state survived. It is useless for computation.",{"type":21,"tag":59,"props":11623,"children":11625},{"id":11624},"the-decoder-families-and-their-bargains",[11626],{"type":26,"value":11627},"The decoder families, and their bargains",{"type":21,"tag":22,"props":11629,"children":11630},{},[11631],{"type":26,"value":11632},"The decoding problem has a beautiful structure in the surface code, and a much uglier one elsewhere.",{"type":21,"tag":22,"props":11634,"children":11635},{},[11636,11641],{"type":21,"tag":71,"props":11637,"children":11638},{},[11639],{"type":26,"value":11640},"Minimum-weight perfect matching (MWPM).",{"type":26,"value":11642}," In the surface code, errors have a geometric signature: a chain of errors flips exactly the two checks at its endpoints. Syndrome violations therefore come in pairs, and finding the most likely error reduces to pairing up the flipped checks so that the total length of the connecting paths is minimised. That's a classic graph problem with a classic polynomial-time solution, Edmonds' blossom algorithm. MWPM has long been the accuracy benchmark for surface-code decoding: roughly a 0.94% threshold on weighted decoder graphs in commonly cited comparisons. Its problem is speed: polynomial time is not the same as fast enough, and the constant factors are unfriendly at microsecond deadlines.",{"type":21,"tag":22,"props":11644,"children":11645},{},[11646,11651],{"type":21,"tag":71,"props":11647,"children":11648},{},[11649],{"type":26,"value":11650},"Union-find.",{"type":26,"value":11652}," The pragmatic answer. Union-find grows clusters around syndrome violations until each cluster is explained by an internal error, using disjoint-set data structures that run in almost-linear time. It's a fast approximation of what matching does, and the accuracy cost is real but modest, with reported thresholds around 0.83% weighted versus MWPM's 0.94%. That trade has made it the workhorse for hardware implementations, particularly FPGA ones.",{"type":21,"tag":22,"props":11654,"children":11655},{},[11656,11661],{"type":21,"tag":71,"props":11657,"children":11658},{},[11659],{"type":26,"value":11660},"Belief propagation plus ordered statistics (BP-OSD).",{"type":26,"value":11662}," This is what you reach for when the matching trick stops working, which is exactly what happens with quantum LDPC codes.",{"type":21,"tag":59,"props":11664,"children":11666},{"id":11665},"why-qldpc-decoding-is-genuinely-harder",[11667],{"type":26,"value":11668},"Why qLDPC decoding is genuinely harder",{"type":21,"tag":22,"props":11670,"children":11671},{},[11672],{"type":26,"value":11673},"The surface code's decoding advantage comes from a property most codes don't share: each error chain flips precisely two checks, so the syndrome maps onto a matching problem. Take that away and you're back to general belief propagation over a Tanner graph: the same message-passing machinery used for classical LDPC codes in modern telecoms.",{"type":21,"tag":22,"props":11675,"children":11676},{},[11677],{"type":26,"value":11678},"Except it works worse in the quantum case, for two structural reasons.",{"type":21,"tag":22,"props":11680,"children":11681},{},[11682,11684,11689],{"type":26,"value":11683},"First, ",{"type":21,"tag":71,"props":11685,"children":11686},{},[11687],{"type":26,"value":11688},"short cycles",{"type":26,"value":11690},". Belief propagation assumes the graph is locally tree-like. The loops in quantum code Tanner graphs break that assumption and hurt convergence.",{"type":21,"tag":22,"props":11692,"children":11693},{},[11694,11696,11701,11703,11708],{"type":26,"value":11695},"Second, and more fundamentally, ",{"type":21,"tag":71,"props":11697,"children":11698},{},[11699],{"type":26,"value":11700},"degeneracy",{"type":26,"value":11702},". In a quantum code, genuinely different physical error patterns are sometimes ",{"type":21,"tag":51,"props":11704,"children":11705},{},[11706],{"type":26,"value":11707},"equivalent",{"type":26,"value":11709},": they differ by a stabilizer and therefore have identical syndromes and identical effects on the logical state. Classically, this doesn't happen. Each syndrome points toward one most likely error. Quantum mechanically, belief propagation gets stuck oscillating between equally valid candidates it has no basis to choose between. It's not that the decoder can't find the answer. It's that there are several answers and the algorithm's tie-breaking machinery wasn't designed for that.",{"type":21,"tag":22,"props":11711,"children":11712},{},[11713],{"type":26,"value":11714},"The standard fix is to bolt ordered statistics decoding onto the back: when BP fails to converge, OSD solves a linear system to pick a candidate. It works, and it costs you. Plain BP scales roughly linearly in code size. The OSD post-processing step is around O(N³). For a decoder on a microsecond budget, cubic scaling is precisely the wrong shape.",{"type":21,"tag":22,"props":11716,"children":11717},{},[11718],{"type":26,"value":11719},"So qLDPC codes present the field with an awkward bargain: better codes, worse decoders.",{"type":21,"tag":59,"props":11721,"children":11723},{"id":11722},"why-anyone-puts-up-with-it",[11724],{"type":26,"value":11725},"Why anyone puts up with it",{"type":21,"tag":22,"props":11727,"children":11728},{},[11729,11731,11736],{"type":26,"value":11730},"Because the encoding rate is dramatically better. Surface codes are extravagant: the ",{"type":21,"tag":29,"props":11732,"children":11733},{"href":1983},[11734],{"type":26,"value":11735},"logical qubit overhead",{"type":26,"value":11737}," runs to hundreds or thousands of physical qubits each, and the fault-tolerance estimates that produce headline figures of millions of physical qubits are mostly surface-code estimates. qLDPC codes, including the bivariate bicycle family, promise far more logical qubits per physical qubit. If that holds up, it moves the timeline for useful fault tolerance by a large margin. That is a prize worth accepting a harder decoding problem for.",{"type":21,"tag":22,"props":11739,"children":11740},{},[11741],{"type":26,"value":11742},"This is where vendors have entered the picture, and where the claims need careful handling.",{"type":21,"tag":22,"props":11744,"children":11745},{},[11746,11751,11753,11758,11760,11765,11767,11772],{"type":21,"tag":71,"props":11747,"children":11748},{},[11749],{"type":26,"value":11750},"IonQ reports",{"type":26,"value":11752}," having developed a \"Beam Search\" decoder intended to replace BP-OSD for quantum LDPC codes. On bivariate bicycle codes, ",{"type":21,"tag":71,"props":11754,"children":11755},{},[11756],{"type":26,"value":11757},"IonQ claims",{"type":26,"value":11759}," a 17x reduction in logical error rate relative to standard BP-OSD, along with a 26x reduction in worst-case (99.9th percentile) runtime, which ",{"type":21,"tag":71,"props":11761,"children":11762},{},[11763],{"type":26,"value":11764},"IonQ states",{"type":26,"value":11766}," it brought under one millisecond on a single core of a commercial CPU. ",{"type":21,"tag":71,"props":11768,"children":11769},{},[11770],{"type":26,"value":11771},"IonQ further estimates",{"type":26,"value":11773}," that three 32-core CPUs would suffice to decode 1,000 logical qubits in its trapped-ion architecture, contrasting that with roughly 1,000 FPGAs for surface-code approaches and 84 FPGAs for superconducting LDPC implementations.",{"type":21,"tag":22,"props":11775,"children":11776},{},[11777,11779,11784],{"type":26,"value":11778},"Those are vendor-reported figures, measured under vendor-chosen conditions, on vendor-selected codes and noise models, and IonQ is a hardware company with a commercial interest in the conclusion that its architecture needs less classical support than its competitors'. The direction of the work is credible and the problem is real. The specific multipliers should be treated as claims pending independent reproduction, in the same way we'd treat any ",{"type":21,"tag":29,"props":11780,"children":11781},{"href":1274},[11782],{"type":26,"value":11783},"benchmark",{"type":26,"value":11785}," that hasn't been replicated by a disinterested party.",{"type":21,"tag":59,"props":11787,"children":11789},{"id":11788},"why-you-cant-simply-pick-the-fast-decoder",[11790],{"type":26,"value":11791},"Why you can't simply pick the fast decoder",{"type":21,"tag":22,"props":11793,"children":11794},{},[11795],{"type":26,"value":11796},"The obvious response to a latency problem is to accept a worse answer faster. In decoding, that instinct is sometimes fatal.",{"type":21,"tag":22,"props":11798,"children":11799},{},[11800,11802,11807,11809,11814,11816,11821,11823,11827],{"type":26,"value":11801},"A less accurate decoder produces a higher logical error rate for the same physical error rate. And the logical error rate is exactly the quantity the threshold theorem is about. ",{"type":21,"tag":29,"props":11803,"children":11804},{"href":4184},[11805],{"type":26,"value":11806},"Google's 2024 below-threshold result",{"type":26,"value":11808}," (the first experimental demonstration that increasing code distance suppresses logical errors rather than amplifying them, descending from ",{"type":21,"tag":29,"props":11810,"children":11811},{"href":9504},[11812],{"type":26,"value":11813},"Kitaev's topological codes",{"type":26,"value":11815}," and ultimately ",{"type":21,"tag":29,"props":11817,"children":11818},{"href":9491},[11819],{"type":26,"value":11820},"Shor's original 1995 code",{"type":26,"value":11822},") was a narrow win. Each distance increase roughly halved the error rate. A decoder that degrades logical ",{"type":21,"tag":29,"props":11824,"children":11825},{"href":9450},[11826],{"type":26,"value":9453},{"type":26,"value":11828}," enough eats that margin entirely and puts you back above threshold, at which point adding qubits makes things worse and the whole exercise inverts.",{"type":21,"tag":22,"props":11830,"children":11831},{},[11832],{"type":26,"value":11833},"So the decoder isn't a peripheral. It sits inside the threshold calculation. You cannot report a threshold crossing without saying which decoder produced it, and a below-threshold result obtained with an offline MWPM decoder is a weaker claim than the same result obtained in real time. Notably, Google's experiment did decode in real time, reporting average decoder latency around 63 µs at distance 5 while sustaining below-threshold performance across up to a million cycles, using an ensemble of neural-network and matching approaches rather than a single clean algorithm.",{"type":21,"tag":59,"props":11835,"children":11837},{"id":11836},"where-the-decoder-physically-lives",[11838],{"type":26,"value":11839},"Where the decoder physically lives",{"type":21,"tag":22,"props":11841,"children":11842},{},[11843,11845,11850],{"type":26,"value":11844},"The last constraint is a plumbing one. Syndrome bits have to travel from the ",{"type":21,"tag":29,"props":11846,"children":11848},{"href":11847},"\u002Fglossary\u002Fqpu",[11849],{"type":26,"value":9326},{"type":26,"value":11851}," to whatever computes on them, and the answer has to travel back, inside the error-correction cycle time.",{"type":21,"tag":22,"props":11853,"children":11854},{},[11855],{"type":26,"value":11856},"That budget covers readout, digitisation, transmission over the classical link, the decode itself, and the return trip. On superconducting hardware the qubits sit in a dilution refrigerator and the classical processor generally doesn't, so the link crosses a temperature boundary. Implementations split roughly three ways: software decoders on CPUs or GPUs (flexible and easy to iterate on, but carrying OS-level latency jitter), FPGA decoders (the current sweet spot for real-time surface-code work), and ASICs (fastest, but freezing your algorithm into silicon in a field where decoding algorithms are still actively improving).",{"type":21,"tag":22,"props":11858,"children":11859},{},[11860,11862,11867,11869,11874],{"type":26,"value":11861},"Slower gate speeds change the maths considerably. Trapped-ion systems operate on microsecond-to-millisecond gate timescales rather than nanoseconds, which relaxes the per-round decoding deadline by orders of magnitude. Part of why a sub-millisecond software decoder is a plausible proposition in that setting and not in a superconducting one. This is another place where ",{"type":21,"tag":29,"props":11863,"children":11864},{"href":814},[11865],{"type":26,"value":11866},"modality differences",{"type":26,"value":11868}," drive architecture rather than being incidental, and it's visible in the ",{"type":21,"tag":29,"props":11870,"children":11871},{"href":862},[11872],{"type":26,"value":11873},"hardware landscape",{"type":26,"value":11875}," if you know to look for it.",{"type":21,"tag":59,"props":11877,"children":11879},{"id":11878},"the-honest-state-of-things",[11880],{"type":26,"value":11881},"The honest state of things",{"type":21,"tag":22,"props":11883,"children":11884},{},[11885],{"type":26,"value":11886},"Real-time decoding is not solved. It is an active, competitive, genuinely open engineering problem, and it scales in an uncomfortable direction: more logical qubits means more syndrome streams, and the classical side has to grow with the quantum side.",{"type":21,"tag":22,"props":11888,"children":11889},{},[11890],{"type":26,"value":11891},"What makes it worth understanding is what it reveals about the shape of the machine. A fault-tolerant quantum computer is not a QPU with a control rack attached. It is a quantum processor and a substantial classical real-time computing system operating as one device, where the classical half's throughput is a hard constraint on whether the quantum half works at all. The decoder is not support equipment. It is part of the computer.",{"type":21,"tag":22,"props":11893,"children":11894},{},[11895,11897,11902],{"type":26,"value":11896},"That framing also explains why progress here is slower to make headlines than qubit counts. There is no satisfying number to announce. But if you're tracking whether fault tolerance is arriving, decoder latency and accuracy under real-time conditions are among the more informative things to watch, more so than most of what appears in ",{"type":21,"tag":29,"props":11898,"children":11899},{"href":10868},[11900],{"type":26,"value":11901},"quantum news",{"type":26,"value":534},{"type":21,"tag":22,"props":11904,"children":11905},{},[11906,11908,11912,11914,11918,11920,11924,11926,11931,11933,11939,11941,11947],{"type":26,"value":11907},"None of this is a reason to wait. The concepts underneath (noise, ",{"type":21,"tag":29,"props":11909,"children":11910},{"href":9366},[11911],{"type":26,"value":9369},{"type":26,"value":11913},", what a ",{"type":21,"tag":29,"props":11915,"children":11916},{"href":1794},[11917],{"type":26,"value":3219},{"type":26,"value":11919}," does under measurement) are the same ones you'd work with on a fault-tolerant machine, and you build that intuition now on ",{"type":21,"tag":29,"props":11921,"children":11922},{"href":1005},[11923],{"type":26,"value":9728},{"type":26,"value":11925},". The ",{"type":21,"tag":29,"props":11927,"children":11928},{"href":11441},[11929],{"type":26,"value":11930},"research library",{"type":26,"value":11932}," has the primary sources, ",{"type":21,"tag":29,"props":11934,"children":11936},{"href":11935},"\u002Fglossary\u002Fsyndrome-decoding",[11937],{"type":26,"value":11938},"syndrome decoding",{"type":26,"value":11940}," has the compact definition, and if you want to see what problems this infrastructure is ultimately being built for, the ",{"type":21,"tag":29,"props":11942,"children":11944},{"href":11943},"\u002Fuse-cases",[11945],{"type":26,"value":11946},"use cases page",{"type":26,"value":11948}," is the place to start.",{"title":7,"searchDepth":335,"depth":335,"links":11950},[11951,11952,11953,11954,11955,11956,11957,11958],{"id":11500,"depth":335,"text":11503},{"id":11555,"depth":335,"text":11558},{"id":11624,"depth":335,"text":11627},{"id":11665,"depth":335,"text":11668},{"id":11722,"depth":335,"text":11725},{"id":11788,"depth":335,"text":11791},{"id":11836,"depth":335,"text":11839},{"id":11878,"depth":335,"text":11881},"content:blog:quantum-error-decoding-bottleneck.md","blog\u002Fquantum-error-decoding-bottleneck.md","blog\u002Fquantum-error-decoding-bottleneck",{"_path":11963,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":11964,"description":11965,"date":3500,"author":11,"tags":11966,"readingTime":3884,"body":11967,"_type":715,"_id":14214,"_source":717,"_file":14215,"_stem":14216,"_extension":720},"\u002Fblog\u002Fquantum-fourier-transform-tutorial","The Quantum Fourier Transform: Build It Yourself in Qiskit","A hands-on tutorial that builds the QFT gate by gate in Qiskit, verifies it against the library implementation, and explains why an exponentially faster Fourier transform gives you no speedup on its own.",[7585,1962,3302],{"type":18,"children":11968,"toc":14203},[11969,11981,11986,11994,12000,12012,12017,12025,12053,12064,12070,12082,12087,12110,12122,12128,12133,12184,12201,12207,12218,12248,12253,12819,12824,12832,12837,12843,12856,13038,13042,13050,13070,13076,13087,13429,13433,13441,13453,13458,13632,13636,13644,13649,13655,13678,13688,13714,13719,13725,13737,14041,14046,14138,14149,14155,14194,14199],{"type":21,"tag":22,"props":11970,"children":11971},{},[11972,11973,11979],{"type":26,"value":11508},{"type":21,"tag":29,"props":11974,"children":11976},{"href":11975},"\u002Fglossary\u002Fqft",[11977],{"type":26,"value":11978},"Quantum Fourier Transform",{"type":26,"value":11980}," is the single most important subroutine in quantum computing. It's the engine inside Shor's factoring algorithm, quantum phase estimation, and order finding. It's also the most commonly misunderstood algorithm in the field, because on paper it looks like an exponentially faster FFT, and it absolutely is not.",{"type":21,"tag":22,"props":11982,"children":11983},{},[11984],{"type":26,"value":11985},"This tutorial builds the QFT from scratch in Qiskit, proves the implementation is correct against Qiskit's own, and then explains carefully where the speedup goes.",{"type":21,"tag":22,"props":11987,"children":11988},{},[11989],{"type":21,"tag":71,"props":11990,"children":11991},{},[11992],{"type":26,"value":11993},"All code in this post was executed against Qiskit 2.5.0 and Qiskit Aer 0.17.2 on Python 3.11. The outputs shown are real.",{"type":21,"tag":59,"props":11995,"children":11997},{"id":11996},"what-the-qft-does",[11998],{"type":26,"value":11999},"What the QFT does",{"type":21,"tag":22,"props":12001,"children":12002},{},[12003,12005,12010],{"type":26,"value":12004},"The classical Discrete Fourier Transform takes a vector of N numbers and returns N Fourier coefficients. The QFT does exactly the same linear map, but the input vector is the ",{"type":21,"tag":51,"props":12006,"children":12007},{},[12008],{"type":26,"value":12009},"amplitude vector",{"type":26,"value":12011}," of an n-qubit state, so N = 2ⁿ.",{"type":21,"tag":22,"props":12013,"children":12014},{},[12015],{"type":26,"value":12016},"On basis states it acts as:",{"type":21,"tag":312,"props":12018,"children":12020},{"code":12019},"QFT |x⟩ = (1\u002F√N) Σ_{k=0}^{N-1} e^{2πi·xk\u002FN} |k⟩\n",[12021],{"type":21,"tag":79,"props":12022,"children":12023},{"__ignoreMap":7},[12024],{"type":26,"value":12019},{"type":21,"tag":22,"props":12026,"children":12027},{},[12028,12030,12035,12037,12043,12045,12051],{"type":26,"value":12029},"Notice what's on the right: every basis state gets amplitude of the ",{"type":21,"tag":51,"props":12031,"children":12032},{},[12033],{"type":26,"value":12034},"same magnitude",{"type":26,"value":12036}," (1\u002F√N), and all the information about ",{"type":21,"tag":79,"props":12038,"children":12040},{"className":12039},[],[12041],{"type":26,"value":12042},"x",{"type":26,"value":12044}," lives in the complex phase ",{"type":21,"tag":79,"props":12046,"children":12048},{"className":12047},[],[12049],{"type":26,"value":12050},"e^{2πi·xk\u002FN}",{"type":26,"value":12052},". The QFT takes information that was sitting in the computational basis and pushes it into the phases.",{"type":21,"tag":22,"props":12054,"children":12055},{},[12056,12058,12062],{"type":26,"value":12057},"That single fact is the whole story of the QFT. It's a perfectly uniform state, as far as any ",{"type":21,"tag":29,"props":12059,"children":12060},{"href":7857},[12061],{"type":26,"value":11538},{"type":26,"value":12063}," in the computational basis is concerned. If you QFT a basis state and measure, you get a uniformly random n-bit string. Every time. You learn nothing.",{"type":21,"tag":59,"props":12065,"children":12067},{"id":12066},"the-readout-problem-stated-plainly",[12068],{"type":26,"value":12069},"The readout problem, stated plainly",{"type":21,"tag":22,"props":12071,"children":12072},{},[12073,12075,12080],{"type":26,"value":12074},"Here's the trap. The QFT on n qubits takes O(n²) gates. Classically, an FFT on the same 2ⁿ amplitudes takes O(n·2ⁿ) operations. That is a genuine exponential gap in ",{"type":21,"tag":51,"props":12076,"children":12077},{},[12078],{"type":26,"value":12079},"gate count",{"type":26,"value":12081},", and it's the number everybody quotes.",{"type":21,"tag":22,"props":12083,"children":12084},{},[12085],{"type":26,"value":12086},"But it isn't a usable speedup, for two reasons:",{"type":21,"tag":7444,"props":12088,"children":12089},{},[12090,12100],{"type":21,"tag":3560,"props":12091,"children":12092},{},[12093,12098],{"type":21,"tag":71,"props":12094,"children":12095},{},[12096],{"type":26,"value":12097},"You can't load the input.",{"type":26,"value":12099}," Getting 2ⁿ arbitrary classical amplitudes into a quantum register generally costs O(2ⁿ) gates, wiping out the advantage before you start.",{"type":21,"tag":3560,"props":12101,"children":12102},{},[12103,12108],{"type":21,"tag":71,"props":12104,"children":12105},{},[12106],{"type":26,"value":12107},"You can't read the output.",{"type":26,"value":12109}," The Fourier coefficients are amplitudes. You cannot look at amplitudes. You only sample from their squared magnitudes, one n-bit string per shot. Extracting all 2ⁿ coefficients would take exponentially many repetitions.",{"type":21,"tag":22,"props":12111,"children":12112},{},[12113,12115,12120],{"type":26,"value":12114},"So the QFT is never a drop-in replacement for the FFT. It is a ",{"type":21,"tag":51,"props":12116,"children":12117},{},[12118],{"type":26,"value":12119},"transform you apply in the middle of a larger circuit",{"type":26,"value":12121},", where the goal isn't to read the spectrum but to convert a periodicity that was hidden in phases into a peak you sample directly. That's the shape every real application takes.",{"type":21,"tag":59,"props":12123,"children":12125},{"id":12124},"the-circuit-hadamard-rotate-recurse-swap",[12126],{"type":26,"value":12127},"The circuit: Hadamard, rotate, recurse, swap",{"type":21,"tag":22,"props":12129,"children":12130},{},[12131],{"type":26,"value":12132},"The construction follows directly from factoring the exponential above into a product over qubits. For the most significant qubit:",{"type":21,"tag":7444,"props":12134,"children":12135},{},[12136,12153,12174,12179],{"type":21,"tag":3560,"props":12137,"children":12138},{},[12139,12140,12145,12147,12151],{"type":26,"value":7689},{"type":21,"tag":29,"props":12141,"children":12142},{"href":7692},[12143],{"type":26,"value":12144},"Hadamard",{"type":26,"value":12146},". This creates a ",{"type":21,"tag":29,"props":12148,"children":12149},{"href":9290},[12150],{"type":26,"value":9293},{"type":26,"value":12152}," whose relative phase encodes one bit of the input.",{"type":21,"tag":3560,"props":12154,"children":12155},{},[12156,12158,12164,12166,12172],{"type":26,"value":12157},"Apply controlled-phase gates ",{"type":21,"tag":79,"props":12159,"children":12161},{"className":12160},[],[12162],{"type":26,"value":12163},"CP(π\u002F2^k)",{"type":26,"value":12165}," from each less significant qubit, controlled onto that same qubit. Qubit distance ",{"type":21,"tag":79,"props":12167,"children":12169},{"className":12168},[],[12170],{"type":26,"value":12171},"k",{"type":26,"value":12173}," contributes a rotation of π\u002F2^k: the further away, the smaller the correction.",{"type":21,"tag":3560,"props":12175,"children":12176},{},[12177],{"type":26,"value":12178},"Recurse on the remaining n−1 qubits.",{"type":21,"tag":3560,"props":12180,"children":12181},{},[12182],{"type":26,"value":12183},"Swap the qubit order at the end.",{"type":21,"tag":22,"props":12185,"children":12186},{},[12187,12192,12194,12199],{"type":21,"tag":71,"props":12188,"children":12189},{},[12190],{"type":26,"value":12191},"Why the swaps?",{"type":26,"value":12193}," The recursion naturally produces the output register in reverse bit order. Qubit 0 ends up holding what should be the most significant output bit, and vice versa. This is a real bug, not a bookkeeping convention. If you feed an unswapped QFT into phase estimation you get a bit-reversed answer. The swaps fix it. (In hardware-aware compilation you sometimes ",{"type":21,"tag":51,"props":12195,"children":12196},{},[12197],{"type":26,"value":12198},"do",{"type":26,"value":12200}," drop the swaps and relabel the wires downstream instead, which is free. But if your QFT is a standalone reusable block, keep them.)",{"type":21,"tag":59,"props":12202,"children":12204},{"id":12203},"building-it-in-qiskit",[12205],{"type":26,"value":12206},"Building it in Qiskit",{"type":21,"tag":22,"props":12208,"children":12209},{},[12210,12212,12216],{"type":26,"value":12211},"Make sure you have ",{"type":21,"tag":29,"props":12213,"children":12214},{"href":7528},[12215],{"type":26,"value":1962},{"type":26,"value":12217}," and the Aer simulator installed:",{"type":21,"tag":312,"props":12219,"children":12221},{"code":12220,"language":10183,"meta":7,"className":10181,"style":7},"pip install qiskit qiskit-aer\n",[12222],{"type":21,"tag":79,"props":12223,"children":12224},{"__ignoreMap":7},[12225],{"type":21,"tag":322,"props":12226,"children":12227},{"class":324,"line":325},[12228,12233,12238,12243],{"type":21,"tag":322,"props":12229,"children":12230},{"style":4801},[12231],{"type":26,"value":12232},"pip",{"type":21,"tag":322,"props":12234,"children":12235},{"style":359},[12236],{"type":26,"value":12237}," install",{"type":21,"tag":322,"props":12239,"children":12240},{"style":359},[12241],{"type":26,"value":12242}," qiskit",{"type":21,"tag":322,"props":12244,"children":12245},{"style":359},[12246],{"type":26,"value":12247}," qiskit-aer\n",{"type":21,"tag":22,"props":12249,"children":12250},{},[12251],{"type":26,"value":12252},"Now the construction, written recursively so it mirrors the math:",{"type":21,"tag":312,"props":12254,"children":12256},{"code":12255,"language":3628,"meta":7,"className":3626,"style":7},"import numpy as np\nfrom qiskit import QuantumCircuit, transpile\nfrom qiskit.quantum_info import Statevector, Operator\nfrom qiskit_aer import AerSimulator\n\n\ndef qft_rotations(circuit, n):\n    \"\"\"Hadamard + controlled phases on the top qubit, then recurse.\"\"\"\n    if n == 0:\n        return circuit\n    n -= 1                      # index of the most significant qubit\n    circuit.h(n)\n    for qubit in range(n):\n        # qubit is (n - qubit) positions below the target\n        circuit.cp(np.pi \u002F 2 ** (n - qubit), qubit, n)\n    qft_rotations(circuit, n)   # recurse on the rest\n    return circuit\n\n\ndef swap_registers(circuit, n):\n    \"\"\"Reverse the qubit order to undo the bit reversal.\"\"\"\n    for qubit in range(n \u002F\u002F 2):\n        circuit.swap(qubit, n - qubit - 1)\n    return circuit\n\n\ndef qft(n):\n    qc = QuantumCircuit(n, name=\"QFT\")\n    qft_rotations(qc, n)\n    swap_registers(qc, n)\n    return qc\n\n\nmy_qft = qft(4)\nprint(my_qft.draw(output=\"text\"))\nprint(my_qft.count_ops())\n",[12257],{"type":21,"tag":79,"props":12258,"children":12259},{"__ignoreMap":7},[12260,12279,12298,12319,12338,12345,12352,12369,12377,12403,12416,12438,12446,12470,12478,12512,12525,12536,12543,12550,12566,12574,12609,12637,12648,12655,12662,12678,12710,12718,12726,12737,12744,12751,12776,12807],{"type":21,"tag":322,"props":12261,"children":12262},{"class":324,"line":325},[12263,12267,12271,12275],{"type":21,"tag":322,"props":12264,"children":12265},{"style":3638},[12266],{"type":26,"value":3651},{"type":21,"tag":322,"props":12268,"children":12269},{"style":329},[12270],{"type":26,"value":4710},{"type":21,"tag":322,"props":12272,"children":12273},{"style":3638},[12274],{"type":26,"value":4715},{"type":21,"tag":322,"props":12276,"children":12277},{"style":329},[12278],{"type":26,"value":4720},{"type":21,"tag":322,"props":12280,"children":12281},{"class":324,"line":335},[12282,12286,12290,12294],{"type":21,"tag":322,"props":12283,"children":12284},{"style":3638},[12285],{"type":26,"value":3641},{"type":21,"tag":322,"props":12287,"children":12288},{"style":329},[12289],{"type":26,"value":4732},{"type":21,"tag":322,"props":12291,"children":12292},{"style":3638},[12293],{"type":26,"value":3651},{"type":21,"tag":322,"props":12295,"children":12296},{"style":329},[12297],{"type":26,"value":4741},{"type":21,"tag":322,"props":12299,"children":12300},{"class":324,"line":344},[12301,12305,12310,12314],{"type":21,"tag":322,"props":12302,"children":12303},{"style":3638},[12304],{"type":26,"value":3641},{"type":21,"tag":322,"props":12306,"children":12307},{"style":329},[12308],{"type":26,"value":12309}," qiskit.quantum_info ",{"type":21,"tag":322,"props":12311,"children":12312},{"style":3638},[12313],{"type":26,"value":3651},{"type":21,"tag":322,"props":12315,"children":12316},{"style":329},[12317],{"type":26,"value":12318}," Statevector, Operator\n",{"type":21,"tag":322,"props":12320,"children":12321},{"class":324,"line":407},[12322,12326,12330,12334],{"type":21,"tag":322,"props":12323,"children":12324},{"style":3638},[12325],{"type":26,"value":3641},{"type":21,"tag":322,"props":12327,"children":12328},{"style":329},[12329],{"type":26,"value":4753},{"type":21,"tag":322,"props":12331,"children":12332},{"style":3638},[12333],{"type":26,"value":3651},{"type":21,"tag":322,"props":12335,"children":12336},{"style":329},[12337],{"type":26,"value":4762},{"type":21,"tag":322,"props":12339,"children":12340},{"class":324,"line":431},[12341],{"type":21,"tag":322,"props":12342,"children":12343},{"emptyLinePlaceholder":3662},[12344],{"type":26,"value":3665},{"type":21,"tag":322,"props":12346,"children":12347},{"class":324,"line":453},[12348],{"type":21,"tag":322,"props":12349,"children":12350},{"emptyLinePlaceholder":3662},[12351],{"type":26,"value":3665},{"type":21,"tag":322,"props":12353,"children":12354},{"class":324,"line":474},[12355,12359,12364],{"type":21,"tag":322,"props":12356,"children":12357},{"style":3638},[12358],{"type":26,"value":4798},{"type":21,"tag":322,"props":12360,"children":12361},{"style":4801},[12362],{"type":26,"value":12363}," qft_rotations",{"type":21,"tag":322,"props":12365,"children":12366},{"style":329},[12367],{"type":26,"value":12368},"(circuit, n):\n",{"type":21,"tag":322,"props":12370,"children":12371},{"class":324,"line":496},[12372],{"type":21,"tag":322,"props":12373,"children":12374},{"style":359},[12375],{"type":26,"value":12376},"    \"\"\"Hadamard + controlled phases on the top qubit, then recurse.\"\"\"\n",{"type":21,"tag":322,"props":12378,"children":12379},{"class":324,"line":16},[12380,12385,12390,12394,12399],{"type":21,"tag":322,"props":12381,"children":12382},{"style":3638},[12383],{"type":26,"value":12384},"    if",{"type":21,"tag":322,"props":12386,"children":12387},{"style":329},[12388],{"type":26,"value":12389}," n ",{"type":21,"tag":322,"props":12391,"children":12392},{"style":3638},[12393],{"type":26,"value":5346},{"type":21,"tag":322,"props":12395,"children":12396},{"style":348},[12397],{"type":26,"value":12398}," 0",{"type":21,"tag":322,"props":12400,"children":12401},{"style":329},[12402],{"type":26,"value":5356},{"type":21,"tag":322,"props":12404,"children":12405},{"class":324,"line":3884},[12406,12411],{"type":21,"tag":322,"props":12407,"children":12408},{"style":3638},[12409],{"type":26,"value":12410},"        return",{"type":21,"tag":322,"props":12412,"children":12413},{"style":329},[12414],{"type":26,"value":12415}," circuit\n",{"type":21,"tag":322,"props":12417,"children":12418},{"class":324,"line":3894},[12419,12424,12429,12433],{"type":21,"tag":322,"props":12420,"children":12421},{"style":329},[12422],{"type":26,"value":12423},"    n ",{"type":21,"tag":322,"props":12425,"children":12426},{"style":3638},[12427],{"type":26,"value":12428},"-=",{"type":21,"tag":322,"props":12430,"children":12431},{"style":348},[12432],{"type":26,"value":6536},{"type":21,"tag":322,"props":12434,"children":12435},{"style":3888},[12436],{"type":26,"value":12437},"                      # index of the most significant qubit\n",{"type":21,"tag":322,"props":12439,"children":12440},{"class":324,"line":1034},[12441],{"type":21,"tag":322,"props":12442,"children":12443},{"style":329},[12444],{"type":26,"value":12445},"    circuit.h(n)\n",{"type":21,"tag":322,"props":12447,"children":12448},{"class":324,"line":3954},[12449,12453,12458,12462,12466],{"type":21,"tag":322,"props":12450,"children":12451},{"style":3638},[12452],{"type":26,"value":5177},{"type":21,"tag":322,"props":12454,"children":12455},{"style":329},[12456],{"type":26,"value":12457}," qubit ",{"type":21,"tag":322,"props":12459,"children":12460},{"style":3638},[12461],{"type":26,"value":3941},{"type":21,"tag":322,"props":12463,"children":12464},{"style":348},[12465],{"type":26,"value":3946},{"type":21,"tag":322,"props":12467,"children":12468},{"style":329},[12469],{"type":26,"value":8278},{"type":21,"tag":322,"props":12471,"children":12472},{"class":324,"line":4002},[12473],{"type":21,"tag":322,"props":12474,"children":12475},{"style":3888},[12476],{"type":26,"value":12477},"        # qubit is (n - qubit) positions below the target\n",{"type":21,"tag":322,"props":12479,"children":12480},{"class":324,"line":4044},[12481,12486,12490,12494,12498,12503,12507],{"type":21,"tag":322,"props":12482,"children":12483},{"style":329},[12484],{"type":26,"value":12485},"        circuit.cp(np.pi ",{"type":21,"tag":322,"props":12487,"children":12488},{"style":3638},[12489],{"type":26,"value":174},{"type":21,"tag":322,"props":12491,"children":12492},{"style":348},[12493],{"type":26,"value":6821},{"type":21,"tag":322,"props":12495,"children":12496},{"style":3638},[12497],{"type":26,"value":8746},{"type":21,"tag":322,"props":12499,"children":12500},{"style":329},[12501],{"type":26,"value":12502}," (n ",{"type":21,"tag":322,"props":12504,"children":12505},{"style":3638},[12506],{"type":26,"value":5285},{"type":21,"tag":322,"props":12508,"children":12509},{"style":329},[12510],{"type":26,"value":12511}," qubit), qubit, n)\n",{"type":21,"tag":322,"props":12513,"children":12514},{"class":324,"line":5068},[12515,12520],{"type":21,"tag":322,"props":12516,"children":12517},{"style":329},[12518],{"type":26,"value":12519},"    qft_rotations(circuit, n)   ",{"type":21,"tag":322,"props":12521,"children":12522},{"style":3888},[12523],{"type":26,"value":12524},"# recurse on the rest\n",{"type":21,"tag":322,"props":12526,"children":12527},{"class":324,"line":5087},[12528,12532],{"type":21,"tag":322,"props":12529,"children":12530},{"style":3638},[12531],{"type":26,"value":5023},{"type":21,"tag":322,"props":12533,"children":12534},{"style":329},[12535],{"type":26,"value":12415},{"type":21,"tag":322,"props":12537,"children":12538},{"class":324,"line":5095},[12539],{"type":21,"tag":322,"props":12540,"children":12541},{"emptyLinePlaceholder":3662},[12542],{"type":26,"value":3665},{"type":21,"tag":322,"props":12544,"children":12545},{"class":324,"line":5126},[12546],{"type":21,"tag":322,"props":12547,"children":12548},{"emptyLinePlaceholder":3662},[12549],{"type":26,"value":3665},{"type":21,"tag":322,"props":12551,"children":12552},{"class":324,"line":5171},[12553,12557,12562],{"type":21,"tag":322,"props":12554,"children":12555},{"style":3638},[12556],{"type":26,"value":4798},{"type":21,"tag":322,"props":12558,"children":12559},{"style":4801},[12560],{"type":26,"value":12561}," swap_registers",{"type":21,"tag":322,"props":12563,"children":12564},{"style":329},[12565],{"type":26,"value":12368},{"type":21,"tag":322,"props":12567,"children":12568},{"class":324,"line":5210},[12569],{"type":21,"tag":322,"props":12570,"children":12571},{"style":359},[12572],{"type":26,"value":12573},"    \"\"\"Reverse the qubit order to undo the bit reversal.\"\"\"\n",{"type":21,"tag":322,"props":12575,"children":12576},{"class":324,"line":5228},[12577,12581,12585,12589,12593,12597,12601,12605],{"type":21,"tag":322,"props":12578,"children":12579},{"style":3638},[12580],{"type":26,"value":5177},{"type":21,"tag":322,"props":12582,"children":12583},{"style":329},[12584],{"type":26,"value":12457},{"type":21,"tag":322,"props":12586,"children":12587},{"style":3638},[12588],{"type":26,"value":3941},{"type":21,"tag":322,"props":12590,"children":12591},{"style":348},[12592],{"type":26,"value":3946},{"type":21,"tag":322,"props":12594,"children":12595},{"style":329},[12596],{"type":26,"value":8106},{"type":21,"tag":322,"props":12598,"children":12599},{"style":3638},[12600],{"type":26,"value":6545},{"type":21,"tag":322,"props":12602,"children":12603},{"style":348},[12604],{"type":26,"value":6821},{"type":21,"tag":322,"props":12606,"children":12607},{"style":329},[12608],{"type":26,"value":4851},{"type":21,"tag":322,"props":12610,"children":12611},{"class":324,"line":5302},[12612,12617,12621,12625,12629,12633],{"type":21,"tag":322,"props":12613,"children":12614},{"style":329},[12615],{"type":26,"value":12616},"        circuit.swap(qubit, n ",{"type":21,"tag":322,"props":12618,"children":12619},{"style":3638},[12620],{"type":26,"value":5285},{"type":21,"tag":322,"props":12622,"children":12623},{"style":329},[12624],{"type":26,"value":12457},{"type":21,"tag":322,"props":12626,"children":12627},{"style":3638},[12628],{"type":26,"value":5285},{"type":21,"tag":322,"props":12630,"children":12631},{"style":348},[12632],{"type":26,"value":6536},{"type":21,"tag":322,"props":12634,"children":12635},{"style":329},[12636],{"type":26,"value":3738},{"type":21,"tag":322,"props":12638,"children":12639},{"class":324,"line":5330},[12640,12644],{"type":21,"tag":322,"props":12641,"children":12642},{"style":3638},[12643],{"type":26,"value":5023},{"type":21,"tag":322,"props":12645,"children":12646},{"style":329},[12647],{"type":26,"value":12415},{"type":21,"tag":322,"props":12649,"children":12650},{"class":324,"line":5359},[12651],{"type":21,"tag":322,"props":12652,"children":12653},{"emptyLinePlaceholder":3662},[12654],{"type":26,"value":3665},{"type":21,"tag":322,"props":12656,"children":12657},{"class":324,"line":5368},[12658],{"type":21,"tag":322,"props":12659,"children":12660},{"emptyLinePlaceholder":3662},[12661],{"type":26,"value":3665},{"type":21,"tag":322,"props":12663,"children":12664},{"class":324,"line":5396},[12665,12669,12674],{"type":21,"tag":322,"props":12666,"children":12667},{"style":3638},[12668],{"type":26,"value":4798},{"type":21,"tag":322,"props":12670,"children":12671},{"style":4801},[12672],{"type":26,"value":12673}," qft",{"type":21,"tag":322,"props":12675,"children":12676},{"style":329},[12677],{"type":26,"value":8278},{"type":21,"tag":322,"props":12679,"children":12680},{"class":324,"line":5427},[12681,12685,12689,12693,12697,12701,12706],{"type":21,"tag":322,"props":12682,"children":12683},{"style":329},[12684],{"type":26,"value":7934},{"type":21,"tag":322,"props":12686,"children":12687},{"style":3638},[12688],{"type":26,"value":3678},{"type":21,"tag":322,"props":12690,"children":12691},{"style":329},[12692],{"type":26,"value":7943},{"type":21,"tag":322,"props":12694,"children":12695},{"style":3703},[12696],{"type":26,"value":7948},{"type":21,"tag":322,"props":12698,"children":12699},{"style":3638},[12700],{"type":26,"value":3678},{"type":21,"tag":322,"props":12702,"children":12703},{"style":359},[12704],{"type":26,"value":12705},"\"QFT\"",{"type":21,"tag":322,"props":12707,"children":12708},{"style":329},[12709],{"type":26,"value":3738},{"type":21,"tag":322,"props":12711,"children":12712},{"class":324,"line":5449},[12713],{"type":21,"tag":322,"props":12714,"children":12715},{"style":329},[12716],{"type":26,"value":12717},"    qft_rotations(qc, n)\n",{"type":21,"tag":322,"props":12719,"children":12720},{"class":324,"line":5495},[12721],{"type":21,"tag":322,"props":12722,"children":12723},{"style":329},[12724],{"type":26,"value":12725},"    swap_registers(qc, n)\n",{"type":21,"tag":322,"props":12727,"children":12728},{"class":324,"line":5508},[12729,12733],{"type":21,"tag":322,"props":12730,"children":12731},{"style":3638},[12732],{"type":26,"value":5023},{"type":21,"tag":322,"props":12734,"children":12735},{"style":329},[12736],{"type":26,"value":8215},{"type":21,"tag":322,"props":12738,"children":12739},{"class":324,"line":5516},[12740],{"type":21,"tag":322,"props":12741,"children":12742},{"emptyLinePlaceholder":3662},[12743],{"type":26,"value":3665},{"type":21,"tag":322,"props":12745,"children":12746},{"class":324,"line":5534},[12747],{"type":21,"tag":322,"props":12748,"children":12749},{"emptyLinePlaceholder":3662},[12750],{"type":26,"value":3665},{"type":21,"tag":322,"props":12752,"children":12753},{"class":324,"line":5552},[12754,12759,12763,12768,12772],{"type":21,"tag":322,"props":12755,"children":12756},{"style":329},[12757],{"type":26,"value":12758},"my_qft ",{"type":21,"tag":322,"props":12760,"children":12761},{"style":3638},[12762],{"type":26,"value":3678},{"type":21,"tag":322,"props":12764,"children":12765},{"style":329},[12766],{"type":26,"value":12767}," qft(",{"type":21,"tag":322,"props":12769,"children":12770},{"style":348},[12771],{"type":26,"value":5969},{"type":21,"tag":322,"props":12773,"children":12774},{"style":329},[12775],{"type":26,"value":3738},{"type":21,"tag":322,"props":12777,"children":12778},{"class":324,"line":5560},[12779,12783,12788,12793,12797,12802],{"type":21,"tag":322,"props":12780,"children":12781},{"style":348},[12782],{"type":26,"value":3753},{"type":21,"tag":322,"props":12784,"children":12785},{"style":329},[12786],{"type":26,"value":12787},"(my_qft.draw(",{"type":21,"tag":322,"props":12789,"children":12790},{"style":3703},[12791],{"type":26,"value":12792},"output",{"type":21,"tag":322,"props":12794,"children":12795},{"style":3638},[12796],{"type":26,"value":3678},{"type":21,"tag":322,"props":12798,"children":12799},{"style":359},[12800],{"type":26,"value":12801},"\"text\"",{"type":21,"tag":322,"props":12803,"children":12804},{"style":329},[12805],{"type":26,"value":12806},"))\n",{"type":21,"tag":322,"props":12808,"children":12809},{"class":324,"line":5594},[12810,12814],{"type":21,"tag":322,"props":12811,"children":12812},{"style":348},[12813],{"type":26,"value":3753},{"type":21,"tag":322,"props":12815,"children":12816},{"style":329},[12817],{"type":26,"value":12818},"(my_qft.count_ops())\n",{"type":21,"tag":22,"props":12820,"children":12821},{},[12822],{"type":26,"value":12823},"Real output:",{"type":21,"tag":312,"props":12825,"children":12827},{"code":12826},"                                                                          ┌───┐»\nq_0: ──────■───────────────────────────────■──────────────────────■───────┤ H ├»\n           │                               │                ┌───┐ │P(π\u002F2) └───┘»\nq_1: ──────┼────────■──────────────────────┼────────■───────┤ H ├─■─────────X──»\n           │        │                ┌───┐ │P(π\u002F4)  │P(π\u002F2) └───┘           │  »\nq_2: ──────┼────────┼────────■───────┤ H ├─■────────■───────────────────────X──»\n     ┌───┐ │P(π\u002F8)  │P(π\u002F4)  │P(π\u002F2) └───┘                                     »\nq_3: ┤ H ├─■────────■────────■─────────────────────────────────────────────────»\n     └───┘\n\nOrderedDict([('cp', 6), ('h', 4), ('swap', 2)])\n",[12828],{"type":21,"tag":79,"props":12829,"children":12830},{"__ignoreMap":7},[12831],{"type":26,"value":12826},{"type":21,"tag":22,"props":12833,"children":12834},{},[12835],{"type":26,"value":12836},"Four Hadamards, six controlled-phase gates, two swaps. That matches the count: n Hadamards and n(n−1)\u002F2 = 6 rotations. The angles descend π\u002F2, π\u002F4, π\u002F8 as the control qubit gets further from the target, exactly as expected.",{"type":21,"tag":59,"props":12838,"children":12840},{"id":12839},"proving-its-correct",[12841],{"type":26,"value":12842},"Proving it's correct",{"type":21,"tag":22,"props":12844,"children":12845},{},[12846,12848,12854],{"type":26,"value":12847},"Never trust a hand-built QFT. Compare the full unitary against Qiskit's ",{"type":21,"tag":79,"props":12849,"children":12851},{"className":12850},[],[12852],{"type":26,"value":12853},"QFTGate",{"type":26,"value":12855},":",{"type":21,"tag":312,"props":12857,"children":12859},{"code":12858,"language":3628,"meta":7,"className":3626,"style":7},"from qiskit.circuit.library import QFTGate\n\nn = 4\nmine = qft(n)\n\nreference = QuantumCircuit(n)\nreference.append(QFTGate(n), range(n))\n\na, b = Operator(mine), Operator(reference)\nprint(\"Equivalent:\", a.equiv(b))\nprint(\"Max element difference:\", np.abs(a.data - b.data).max())\n",[12860],{"type":21,"tag":79,"props":12861,"children":12862},{"__ignoreMap":7},[12863,12883,12890,12906,12923,12930,12947,12963,12970,12987,13008],{"type":21,"tag":322,"props":12864,"children":12865},{"class":324,"line":325},[12866,12870,12874,12878],{"type":21,"tag":322,"props":12867,"children":12868},{"style":3638},[12869],{"type":26,"value":3641},{"type":21,"tag":322,"props":12871,"children":12872},{"style":329},[12873],{"type":26,"value":6258},{"type":21,"tag":322,"props":12875,"children":12876},{"style":3638},[12877],{"type":26,"value":3651},{"type":21,"tag":322,"props":12879,"children":12880},{"style":329},[12881],{"type":26,"value":12882}," QFTGate\n",{"type":21,"tag":322,"props":12884,"children":12885},{"class":324,"line":335},[12886],{"type":21,"tag":322,"props":12887,"children":12888},{"emptyLinePlaceholder":3662},[12889],{"type":26,"value":3665},{"type":21,"tag":322,"props":12891,"children":12892},{"class":324,"line":344},[12893,12897,12901],{"type":21,"tag":322,"props":12894,"children":12895},{"style":329},[12896],{"type":26,"value":5522},{"type":21,"tag":322,"props":12898,"children":12899},{"style":3638},[12900],{"type":26,"value":3678},{"type":21,"tag":322,"props":12902,"children":12903},{"style":348},[12904],{"type":26,"value":12905}," 4\n",{"type":21,"tag":322,"props":12907,"children":12908},{"class":324,"line":407},[12909,12914,12918],{"type":21,"tag":322,"props":12910,"children":12911},{"style":329},[12912],{"type":26,"value":12913},"mine ",{"type":21,"tag":322,"props":12915,"children":12916},{"style":3638},[12917],{"type":26,"value":3678},{"type":21,"tag":322,"props":12919,"children":12920},{"style":329},[12921],{"type":26,"value":12922}," qft(n)\n",{"type":21,"tag":322,"props":12924,"children":12925},{"class":324,"line":431},[12926],{"type":21,"tag":322,"props":12927,"children":12928},{"emptyLinePlaceholder":3662},[12929],{"type":26,"value":3665},{"type":21,"tag":322,"props":12931,"children":12932},{"class":324,"line":453},[12933,12938,12942],{"type":21,"tag":322,"props":12934,"children":12935},{"style":329},[12936],{"type":26,"value":12937},"reference ",{"type":21,"tag":322,"props":12939,"children":12940},{"style":3638},[12941],{"type":26,"value":3678},{"type":21,"tag":322,"props":12943,"children":12944},{"style":329},[12945],{"type":26,"value":12946}," QuantumCircuit(n)\n",{"type":21,"tag":322,"props":12948,"children":12949},{"class":324,"line":474},[12950,12955,12959],{"type":21,"tag":322,"props":12951,"children":12952},{"style":329},[12953],{"type":26,"value":12954},"reference.append(QFTGate(n), ",{"type":21,"tag":322,"props":12956,"children":12957},{"style":348},[12958],{"type":26,"value":5379},{"type":21,"tag":322,"props":12960,"children":12961},{"style":329},[12962],{"type":26,"value":5393},{"type":21,"tag":322,"props":12964,"children":12965},{"class":324,"line":496},[12966],{"type":21,"tag":322,"props":12967,"children":12968},{"emptyLinePlaceholder":3662},[12969],{"type":26,"value":3665},{"type":21,"tag":322,"props":12971,"children":12972},{"class":324,"line":16},[12973,12978,12982],{"type":21,"tag":322,"props":12974,"children":12975},{"style":329},[12976],{"type":26,"value":12977},"a, b ",{"type":21,"tag":322,"props":12979,"children":12980},{"style":3638},[12981],{"type":26,"value":3678},{"type":21,"tag":322,"props":12983,"children":12984},{"style":329},[12985],{"type":26,"value":12986}," Operator(mine), Operator(reference)\n",{"type":21,"tag":322,"props":12988,"children":12989},{"class":324,"line":3884},[12990,12994,12998,13003],{"type":21,"tag":322,"props":12991,"children":12992},{"style":348},[12993],{"type":26,"value":3753},{"type":21,"tag":322,"props":12995,"children":12996},{"style":329},[12997],{"type":26,"value":3758},{"type":21,"tag":322,"props":12999,"children":13000},{"style":359},[13001],{"type":26,"value":13002},"\"Equivalent:\"",{"type":21,"tag":322,"props":13004,"children":13005},{"style":329},[13006],{"type":26,"value":13007},", a.equiv(b))\n",{"type":21,"tag":322,"props":13009,"children":13010},{"class":324,"line":3894},[13011,13015,13019,13024,13029,13033],{"type":21,"tag":322,"props":13012,"children":13013},{"style":348},[13014],{"type":26,"value":3753},{"type":21,"tag":322,"props":13016,"children":13017},{"style":329},[13018],{"type":26,"value":3758},{"type":21,"tag":322,"props":13020,"children":13021},{"style":359},[13022],{"type":26,"value":13023},"\"Max element difference:\"",{"type":21,"tag":322,"props":13025,"children":13026},{"style":329},[13027],{"type":26,"value":13028},", np.abs(a.data ",{"type":21,"tag":322,"props":13030,"children":13031},{"style":3638},[13032],{"type":26,"value":5285},{"type":21,"tag":322,"props":13034,"children":13035},{"style":329},[13036],{"type":26,"value":13037}," b.data).max())\n",{"type":21,"tag":22,"props":13039,"children":13040},{},[13041],{"type":26,"value":12823},{"type":21,"tag":312,"props":13043,"children":13045},{"code":13044},"Equivalent: True\nMax element difference: 2.285583597498884e-15\n",[13046],{"type":21,"tag":79,"props":13047,"children":13048},{"__ignoreMap":7},[13049],{"type":26,"value":13044},{"type":21,"tag":22,"props":13051,"children":13052},{},[13053,13055,13061,13062,13068],{"type":26,"value":13054},"Identical to floating-point precision, not merely equal up to a global phase. If you're on an older Qiskit, use ",{"type":21,"tag":79,"props":13056,"children":13058},{"className":13057},[],[13059],{"type":26,"value":13060},"from qiskit.circuit.library import QFT",{"type":26,"value":236},{"type":21,"tag":79,"props":13063,"children":13065},{"className":13064},[],[13066],{"type":26,"value":13067},"QFT(n).decompose()",{"type":26,"value":13069}," instead.",{"type":21,"tag":59,"props":13071,"children":13073},{"id":13072},"seeing-it-find-a-period",[13074],{"type":26,"value":13075},"Seeing it find a period",{"type":21,"tag":22,"props":13077,"children":13078},{},[13079,13081,13085],{"type":26,"value":13080},"This is the one demonstration that shows what the QFT is ",{"type":21,"tag":51,"props":13082,"children":13083},{},[13084],{"type":26,"value":3931},{"type":26,"value":13086},". Prepare a state that's a uniform superposition over multiples of 4 (a comb with period 4) and transform it:",{"type":21,"tag":312,"props":13088,"children":13090},{"code":13089,"language":3628,"meta":7,"className":3626,"style":7},"n = 4\npsi = np.zeros(2 ** n, dtype=complex)\nfor x in range(0, 2 ** n, 4):   # |0> + |4> + |8> + |12>\n    psi[x] = 1\npsi \u002F= np.linalg.norm(psi)\n\nout = Statevector(psi).evolve(qft(n))\nfor i, p in enumerate(out.probabilities()):\n    if p > 1e-6:\n        print(f\"|{i:0{n}b}> = {i:2d}   p = {p:.4f}\")\n",[13091],{"type":21,"tag":79,"props":13092,"children":13093},{"__ignoreMap":7},[13094,13109,13157,13215,13232,13248,13255,13272,13297,13323],{"type":21,"tag":322,"props":13095,"children":13096},{"class":324,"line":325},[13097,13101,13105],{"type":21,"tag":322,"props":13098,"children":13099},{"style":329},[13100],{"type":26,"value":5522},{"type":21,"tag":322,"props":13102,"children":13103},{"style":3638},[13104],{"type":26,"value":3678},{"type":21,"tag":322,"props":13106,"children":13107},{"style":348},[13108],{"type":26,"value":12905},{"type":21,"tag":322,"props":13110,"children":13111},{"class":324,"line":335},[13112,13117,13121,13126,13130,13134,13139,13144,13148,13153],{"type":21,"tag":322,"props":13113,"children":13114},{"style":329},[13115],{"type":26,"value":13116},"psi ",{"type":21,"tag":322,"props":13118,"children":13119},{"style":3638},[13120],{"type":26,"value":3678},{"type":21,"tag":322,"props":13122,"children":13123},{"style":329},[13124],{"type":26,"value":13125}," np.zeros(",{"type":21,"tag":322,"props":13127,"children":13128},{"style":348},[13129],{"type":26,"value":423},{"type":21,"tag":322,"props":13131,"children":13132},{"style":3638},[13133],{"type":26,"value":8746},{"type":21,"tag":322,"props":13135,"children":13136},{"style":329},[13137],{"type":26,"value":13138}," n, ",{"type":21,"tag":322,"props":13140,"children":13141},{"style":3703},[13142],{"type":26,"value":13143},"dtype",{"type":21,"tag":322,"props":13145,"children":13146},{"style":3638},[13147],{"type":26,"value":3678},{"type":21,"tag":322,"props":13149,"children":13150},{"style":348},[13151],{"type":26,"value":13152},"complex",{"type":21,"tag":322,"props":13154,"children":13155},{"style":329},[13156],{"type":26,"value":3738},{"type":21,"tag":322,"props":13158,"children":13159},{"class":324,"line":344},[13160,13164,13169,13173,13177,13181,13185,13189,13193,13197,13201,13205,13210],{"type":21,"tag":322,"props":13161,"children":13162},{"style":3638},[13163],{"type":26,"value":3931},{"type":21,"tag":322,"props":13165,"children":13166},{"style":329},[13167],{"type":26,"value":13168}," x ",{"type":21,"tag":322,"props":13170,"children":13171},{"style":3638},[13172],{"type":26,"value":3941},{"type":21,"tag":322,"props":13174,"children":13175},{"style":348},[13176],{"type":26,"value":3946},{"type":21,"tag":322,"props":13178,"children":13179},{"style":329},[13180],{"type":26,"value":3758},{"type":21,"tag":322,"props":13182,"children":13183},{"style":348},[13184],{"type":26,"value":5605},{"type":21,"tag":322,"props":13186,"children":13187},{"style":329},[13188],{"type":26,"value":367},{"type":21,"tag":322,"props":13190,"children":13191},{"style":348},[13192],{"type":26,"value":423},{"type":21,"tag":322,"props":13194,"children":13195},{"style":3638},[13196],{"type":26,"value":8746},{"type":21,"tag":322,"props":13198,"children":13199},{"style":329},[13200],{"type":26,"value":13138},{"type":21,"tag":322,"props":13202,"children":13203},{"style":348},[13204],{"type":26,"value":5969},{"type":21,"tag":322,"props":13206,"children":13207},{"style":329},[13208],{"type":26,"value":13209},"):   ",{"type":21,"tag":322,"props":13211,"children":13212},{"style":3888},[13213],{"type":26,"value":13214},"# |0> + |4> + |8> + |12>\n",{"type":21,"tag":322,"props":13216,"children":13217},{"class":324,"line":407},[13218,13223,13227],{"type":21,"tag":322,"props":13219,"children":13220},{"style":329},[13221],{"type":26,"value":13222},"    psi[x] ",{"type":21,"tag":322,"props":13224,"children":13225},{"style":3638},[13226],{"type":26,"value":3678},{"type":21,"tag":322,"props":13228,"children":13229},{"style":348},[13230],{"type":26,"value":13231}," 1\n",{"type":21,"tag":322,"props":13233,"children":13234},{"class":324,"line":431},[13235,13239,13243],{"type":21,"tag":322,"props":13236,"children":13237},{"style":329},[13238],{"type":26,"value":13116},{"type":21,"tag":322,"props":13240,"children":13241},{"style":3638},[13242],{"type":26,"value":5901},{"type":21,"tag":322,"props":13244,"children":13245},{"style":329},[13246],{"type":26,"value":13247}," np.linalg.norm(psi)\n",{"type":21,"tag":322,"props":13249,"children":13250},{"class":324,"line":453},[13251],{"type":21,"tag":322,"props":13252,"children":13253},{"emptyLinePlaceholder":3662},[13254],{"type":26,"value":3665},{"type":21,"tag":322,"props":13256,"children":13257},{"class":324,"line":474},[13258,13263,13267],{"type":21,"tag":322,"props":13259,"children":13260},{"style":329},[13261],{"type":26,"value":13262},"out ",{"type":21,"tag":322,"props":13264,"children":13265},{"style":3638},[13266],{"type":26,"value":3678},{"type":21,"tag":322,"props":13268,"children":13269},{"style":329},[13270],{"type":26,"value":13271}," Statevector(psi).evolve(qft(n))\n",{"type":21,"tag":322,"props":13273,"children":13274},{"class":324,"line":496},[13275,13279,13284,13288,13292],{"type":21,"tag":322,"props":13276,"children":13277},{"style":3638},[13278],{"type":26,"value":3931},{"type":21,"tag":322,"props":13280,"children":13281},{"style":329},[13282],{"type":26,"value":13283}," i, p ",{"type":21,"tag":322,"props":13285,"children":13286},{"style":3638},[13287],{"type":26,"value":3941},{"type":21,"tag":322,"props":13289,"children":13290},{"style":348},[13291],{"type":26,"value":5322},{"type":21,"tag":322,"props":13293,"children":13294},{"style":329},[13295],{"type":26,"value":13296},"(out.probabilities()):\n",{"type":21,"tag":322,"props":13298,"children":13299},{"class":324,"line":16},[13300,13304,13309,13314,13319],{"type":21,"tag":322,"props":13301,"children":13302},{"style":3638},[13303],{"type":26,"value":12384},{"type":21,"tag":322,"props":13305,"children":13306},{"style":329},[13307],{"type":26,"value":13308}," p ",{"type":21,"tag":322,"props":13310,"children":13311},{"style":3638},[13312],{"type":26,"value":13313},">",{"type":21,"tag":322,"props":13315,"children":13316},{"style":348},[13317],{"type":26,"value":13318}," 1e-6",{"type":21,"tag":322,"props":13320,"children":13321},{"style":329},[13322],{"type":26,"value":5356},{"type":21,"tag":322,"props":13324,"children":13325},{"class":324,"line":3884},[13326,13331,13335,13339,13344,13348,13352,13357,13361,13365,13369,13374,13378,13383,13387,13391,13396,13400,13405,13409,13413,13417,13421,13425],{"type":21,"tag":322,"props":13327,"children":13328},{"style":348},[13329],{"type":26,"value":13330},"        print",{"type":21,"tag":322,"props":13332,"children":13333},{"style":329},[13334],{"type":26,"value":3758},{"type":21,"tag":322,"props":13336,"children":13337},{"style":3638},[13338],{"type":26,"value":3763},{"type":21,"tag":322,"props":13340,"children":13341},{"style":359},[13342],{"type":26,"value":13343},"\"|",{"type":21,"tag":322,"props":13345,"children":13346},{"style":348},[13347],{"type":26,"value":3773},{"type":21,"tag":322,"props":13349,"children":13350},{"style":329},[13351],{"type":26,"value":4673},{"type":21,"tag":322,"props":13353,"children":13354},{"style":3638},[13355],{"type":26,"value":13356},":0",{"type":21,"tag":322,"props":13358,"children":13359},{"style":348},[13360],{"type":26,"value":3773},{"type":21,"tag":322,"props":13362,"children":13363},{"style":329},[13364],{"type":26,"value":5266},{"type":21,"tag":322,"props":13366,"children":13367},{"style":348},[13368],{"type":26,"value":3783},{"type":21,"tag":322,"props":13370,"children":13371},{"style":3638},[13372],{"type":26,"value":13373},"b",{"type":21,"tag":322,"props":13375,"children":13376},{"style":348},[13377],{"type":26,"value":3783},{"type":21,"tag":322,"props":13379,"children":13380},{"style":359},[13381],{"type":26,"value":13382},"> = ",{"type":21,"tag":322,"props":13384,"children":13385},{"style":348},[13386],{"type":26,"value":3773},{"type":21,"tag":322,"props":13388,"children":13389},{"style":329},[13390],{"type":26,"value":4673},{"type":21,"tag":322,"props":13392,"children":13393},{"style":3638},[13394],{"type":26,"value":13395},":2d",{"type":21,"tag":322,"props":13397,"children":13398},{"style":348},[13399],{"type":26,"value":3783},{"type":21,"tag":322,"props":13401,"children":13402},{"style":359},[13403],{"type":26,"value":13404},"   p = ",{"type":21,"tag":322,"props":13406,"children":13407},{"style":348},[13408],{"type":26,"value":3773},{"type":21,"tag":322,"props":13410,"children":13411},{"style":329},[13412],{"type":26,"value":22},{"type":21,"tag":322,"props":13414,"children":13415},{"style":3638},[13416],{"type":26,"value":4085},{"type":21,"tag":322,"props":13418,"children":13419},{"style":348},[13420],{"type":26,"value":3783},{"type":21,"tag":322,"props":13422,"children":13423},{"style":359},[13424],{"type":26,"value":3788},{"type":21,"tag":322,"props":13426,"children":13427},{"style":329},[13428],{"type":26,"value":3738},{"type":21,"tag":22,"props":13430,"children":13431},{},[13432],{"type":26,"value":12823},{"type":21,"tag":312,"props":13434,"children":13436},{"code":13435},"|0000> =  0   p = 0.2500\n|0100> =  4   p = 0.2500\n|1000> =  8   p = 0.2500\n|1100> = 12   p = 0.2500\n",[13437],{"type":21,"tag":79,"props":13438,"children":13439},{"__ignoreMap":7},[13440],{"type":26,"value":13435},{"type":21,"tag":22,"props":13442,"children":13443},{},[13444,13446,13451],{"type":26,"value":13445},"The input had period 4 in position space. The output is concentrated on multiples of 16\u002F4 = 4 in frequency space. Sample this a few times, take the greatest common divisor of the outcomes, and you recover the period. ",{"type":21,"tag":71,"props":13447,"children":13448},{},[13449],{"type":26,"value":13450},"That",{"type":26,"value":13452}," is the readable signal: not the coefficients, but a periodicity converted into a sampling distribution.",{"type":21,"tag":22,"props":13454,"children":13455},{},[13456],{"type":26,"value":13457},"Confirm it on the simulator with actual shots:",{"type":21,"tag":312,"props":13459,"children":13461},{"code":13460,"language":3628,"meta":7,"className":3626,"style":7},"qc = QuantumCircuit(n)\nqc.initialize(psi, range(n))\nqc.compose(qft(n), inplace=True)\nqc.measure_all()\n\nsim = AerSimulator()\ncounts = sim.run(transpile(qc, sim), shots=4096).result().get_counts()\nprint(sorted(counts.items(), key=lambda kv: -kv[1]))\n",[13462],{"type":21,"tag":79,"props":13463,"children":13464},{"__ignoreMap":7},[13465,13480,13496,13520,13528,13535,13550,13581],{"type":21,"tag":322,"props":13466,"children":13467},{"class":324,"line":325},[13468,13472,13476],{"type":21,"tag":322,"props":13469,"children":13470},{"style":329},[13471],{"type":26,"value":7819},{"type":21,"tag":322,"props":13473,"children":13474},{"style":3638},[13475],{"type":26,"value":3678},{"type":21,"tag":322,"props":13477,"children":13478},{"style":329},[13479],{"type":26,"value":12946},{"type":21,"tag":322,"props":13481,"children":13482},{"class":324,"line":335},[13483,13488,13492],{"type":21,"tag":322,"props":13484,"children":13485},{"style":329},[13486],{"type":26,"value":13487},"qc.initialize(psi, ",{"type":21,"tag":322,"props":13489,"children":13490},{"style":348},[13491],{"type":26,"value":5379},{"type":21,"tag":322,"props":13493,"children":13494},{"style":329},[13495],{"type":26,"value":5393},{"type":21,"tag":322,"props":13497,"children":13498},{"class":324,"line":344},[13499,13504,13508,13512,13516],{"type":21,"tag":322,"props":13500,"children":13501},{"style":329},[13502],{"type":26,"value":13503},"qc.compose(qft(n), ",{"type":21,"tag":322,"props":13505,"children":13506},{"style":3703},[13507],{"type":26,"value":8636},{"type":21,"tag":322,"props":13509,"children":13510},{"style":3638},[13511],{"type":26,"value":3678},{"type":21,"tag":322,"props":13513,"children":13514},{"style":348},[13515],{"type":26,"value":3715},{"type":21,"tag":322,"props":13517,"children":13518},{"style":329},[13519],{"type":26,"value":3738},{"type":21,"tag":322,"props":13521,"children":13522},{"class":324,"line":407},[13523],{"type":21,"tag":322,"props":13524,"children":13525},{"style":329},[13526],{"type":26,"value":13527},"qc.measure_all()\n",{"type":21,"tag":322,"props":13529,"children":13530},{"class":324,"line":431},[13531],{"type":21,"tag":322,"props":13532,"children":13533},{"emptyLinePlaceholder":3662},[13534],{"type":26,"value":3665},{"type":21,"tag":322,"props":13536,"children":13537},{"class":324,"line":453},[13538,13542,13546],{"type":21,"tag":322,"props":13539,"children":13540},{"style":329},[13541],{"type":26,"value":8861},{"type":21,"tag":322,"props":13543,"children":13544},{"style":3638},[13545],{"type":26,"value":3678},{"type":21,"tag":322,"props":13547,"children":13548},{"style":329},[13549],{"type":26,"value":8870},{"type":21,"tag":322,"props":13551,"children":13552},{"class":324,"line":474},[13553,13557,13561,13565,13569,13573,13577],{"type":21,"tag":322,"props":13554,"children":13555},{"style":329},[13556],{"type":26,"value":8894},{"type":21,"tag":322,"props":13558,"children":13559},{"style":3638},[13560],{"type":26,"value":3678},{"type":21,"tag":322,"props":13562,"children":13563},{"style":329},[13564],{"type":26,"value":8903},{"type":21,"tag":322,"props":13566,"children":13567},{"style":3703},[13568],{"type":26,"value":4560},{"type":21,"tag":322,"props":13570,"children":13571},{"style":3638},[13572],{"type":26,"value":3678},{"type":21,"tag":322,"props":13574,"children":13575},{"style":348},[13576],{"type":26,"value":8916},{"type":21,"tag":322,"props":13578,"children":13579},{"style":329},[13580],{"type":26,"value":5719},{"type":21,"tag":322,"props":13582,"children":13583},{"class":324,"line":496},[13584,13588,13592,13596,13601,13605,13609,13614,13618,13623,13627],{"type":21,"tag":322,"props":13585,"children":13586},{"style":348},[13587],{"type":26,"value":3753},{"type":21,"tag":322,"props":13589,"children":13590},{"style":329},[13591],{"type":26,"value":3758},{"type":21,"tag":322,"props":13593,"children":13594},{"style":348},[13595],{"type":26,"value":8936},{"type":21,"tag":322,"props":13597,"children":13598},{"style":329},[13599],{"type":26,"value":13600},"(counts.items(), ",{"type":21,"tag":322,"props":13602,"children":13603},{"style":3703},[13604],{"type":26,"value":3979},{"type":21,"tag":322,"props":13606,"children":13607},{"style":3638},[13608],{"type":26,"value":3984},{"type":21,"tag":322,"props":13610,"children":13611},{"style":329},[13612],{"type":26,"value":13613}," kv: ",{"type":21,"tag":322,"props":13615,"children":13616},{"style":3638},[13617],{"type":26,"value":5285},{"type":21,"tag":322,"props":13619,"children":13620},{"style":329},[13621],{"type":26,"value":13622},"kv[",{"type":21,"tag":322,"props":13624,"children":13625},{"style":348},[13626],{"type":26,"value":3994},{"type":21,"tag":322,"props":13628,"children":13629},{"style":329},[13630],{"type":26,"value":13631},"]))\n",{"type":21,"tag":22,"props":13633,"children":13634},{},[13635],{"type":26,"value":12823},{"type":21,"tag":312,"props":13637,"children":13639},{"code":13638},"[('0100', 1079), ('1000', 1030), ('0000', 1018), ('1100', 969)]\n",[13640],{"type":21,"tag":79,"props":13641,"children":13642},{"__ignoreMap":7},[13643],{"type":26,"value":13638},{"type":21,"tag":22,"props":13645,"children":13646},{},[13647],{"type":26,"value":13648},"Four peaks, roughly 1024 counts each. Nothing else appears.",{"type":21,"tag":59,"props":13650,"children":13652},{"id":13651},"where-the-qft-earns-its-keep",[13653],{"type":26,"value":13654},"Where the QFT earns its keep",{"type":21,"tag":22,"props":13656,"children":13657},{},[13658,13663,13665,13669,13671,13677],{"type":21,"tag":71,"props":13659,"children":13660},{},[13661],{"type":26,"value":13662},"Quantum phase estimation (QPE).",{"type":26,"value":13664}," Given a unitary U and an eigenstate |ψ⟩ with U|ψ⟩ = e^{2πiθ}|ψ⟩, QPE writes θ into a register's phases using controlled-U powers, then applies the ",{"type":21,"tag":51,"props":13666,"children":13667},{},[13668],{"type":26,"value":7416},{"type":26,"value":13670}," QFT to convert those phases into a binary number you measure. Note the direction: QPE uses QFT⁻¹, because the phases are already there and the job is to pull them out. In Qiskit that's ",{"type":21,"tag":79,"props":13672,"children":13674},{"className":13673},[],[13675],{"type":26,"value":13676},"qft(n).inverse()",{"type":26,"value":534},{"type":21,"tag":22,"props":13679,"children":13680},{},[13681,13686],{"type":21,"tag":71,"props":13682,"children":13683},{},[13684],{"type":26,"value":13685},"Order finding.",{"type":26,"value":13687}," Given a and N, find the smallest r with a^r ≡ 1 (mod N). Modular exponentiation creates a register periodic in r. The QFT turns that period into measurable peaks, exactly as in the demo above.",{"type":21,"tag":22,"props":13689,"children":13690},{},[13691,13699,13701,13706,13708,13712],{"type":21,"tag":71,"props":13692,"children":13693},{},[13694,13698],{"type":21,"tag":29,"props":13695,"children":13696},{"href":10066},[13697],{"type":26,"value":10069},{"type":26,"value":534},{"type":26,"value":13700}," Factoring reduces to order finding, so Shor's is order finding plus classical post-processing (continued fractions and a GCD). The ",{"type":21,"tag":29,"props":13702,"children":13703},{"href":9045},[13704],{"type":26,"value":13705},"1994 paper",{"type":26,"value":13707}," is worth reading precisely because the QFT is the only quantum part: everything else is number theory. This is also why the algorithm threatens RSA, and why ",{"type":21,"tag":29,"props":13709,"children":13710},{"href":9817},[13711],{"type":26,"value":9800},{"type":26,"value":13713}," migration is happening now rather than later.",{"type":21,"tag":22,"props":13715,"children":13716},{},[13717],{"type":26,"value":13718},"In all three cases the QFT is doing the same job: turning phase information into position information so a measurement sees it.",{"type":21,"tag":59,"props":13720,"children":13722},{"id":13721},"approximate-qft-drop-the-tiny-rotations",[13723],{"type":26,"value":13724},"Approximate QFT: drop the tiny rotations",{"type":21,"tag":22,"props":13726,"children":13727},{},[13728,13730,13735],{"type":26,"value":13729},"On real hardware, ",{"type":21,"tag":79,"props":13731,"children":13733},{"className":13732},[],[13734],{"type":26,"value":12163},{"type":26,"value":13736}," for large k is a rotation smaller than your gate error. Executing it adds noise and buys nothing. The Approximate QFT (AQFT) simply skips every rotation with k above a cutoff, reducing the gate count from O(n²) to O(n log n):",{"type":21,"tag":312,"props":13738,"children":13740},{"code":13739,"language":3628,"meta":7,"className":3626,"style":7},"def aqft(n, k_max):\n    qc = QuantumCircuit(n, name=\"AQFT\")\n    for size in range(n, 0, -1):\n        m = size - 1\n        qc.h(m)\n        for qubit in range(m):\n            k = m - qubit\n            if k \u003C= k_max:\n                qc.cp(np.pi \u002F 2 ** k, qubit, m)\n    for q in range(n \u002F\u002F 2):\n        qc.swap(q, n - q - 1)\n    return qc\n",[13741],{"type":21,"tag":79,"props":13742,"children":13743},{"__ignoreMap":7},[13744,13761,13793,13838,13862,13870,13894,13920,13942,13967,14002,14030],{"type":21,"tag":322,"props":13745,"children":13746},{"class":324,"line":325},[13747,13751,13756],{"type":21,"tag":322,"props":13748,"children":13749},{"style":3638},[13750],{"type":26,"value":4798},{"type":21,"tag":322,"props":13752,"children":13753},{"style":4801},[13754],{"type":26,"value":13755}," aqft",{"type":21,"tag":322,"props":13757,"children":13758},{"style":329},[13759],{"type":26,"value":13760},"(n, k_max):\n",{"type":21,"tag":322,"props":13762,"children":13763},{"class":324,"line":335},[13764,13768,13772,13776,13780,13784,13789],{"type":21,"tag":322,"props":13765,"children":13766},{"style":329},[13767],{"type":26,"value":7934},{"type":21,"tag":322,"props":13769,"children":13770},{"style":3638},[13771],{"type":26,"value":3678},{"type":21,"tag":322,"props":13773,"children":13774},{"style":329},[13775],{"type":26,"value":7943},{"type":21,"tag":322,"props":13777,"children":13778},{"style":3703},[13779],{"type":26,"value":7948},{"type":21,"tag":322,"props":13781,"children":13782},{"style":3638},[13783],{"type":26,"value":3678},{"type":21,"tag":322,"props":13785,"children":13786},{"style":359},[13787],{"type":26,"value":13788},"\"AQFT\"",{"type":21,"tag":322,"props":13790,"children":13791},{"style":329},[13792],{"type":26,"value":3738},{"type":21,"tag":322,"props":13794,"children":13795},{"class":324,"line":344},[13796,13800,13805,13809,13813,13818,13822,13826,13830,13834],{"type":21,"tag":322,"props":13797,"children":13798},{"style":3638},[13799],{"type":26,"value":5177},{"type":21,"tag":322,"props":13801,"children":13802},{"style":329},[13803],{"type":26,"value":13804}," size ",{"type":21,"tag":322,"props":13806,"children":13807},{"style":3638},[13808],{"type":26,"value":3941},{"type":21,"tag":322,"props":13810,"children":13811},{"style":348},[13812],{"type":26,"value":3946},{"type":21,"tag":322,"props":13814,"children":13815},{"style":329},[13816],{"type":26,"value":13817},"(n, ",{"type":21,"tag":322,"props":13819,"children":13820},{"style":348},[13821],{"type":26,"value":5605},{"type":21,"tag":322,"props":13823,"children":13824},{"style":329},[13825],{"type":26,"value":367},{"type":21,"tag":322,"props":13827,"children":13828},{"style":3638},[13829],{"type":26,"value":5285},{"type":21,"tag":322,"props":13831,"children":13832},{"style":348},[13833],{"type":26,"value":3994},{"type":21,"tag":322,"props":13835,"children":13836},{"style":329},[13837],{"type":26,"value":4851},{"type":21,"tag":322,"props":13839,"children":13840},{"class":324,"line":407},[13841,13846,13850,13854,13858],{"type":21,"tag":322,"props":13842,"children":13843},{"style":329},[13844],{"type":26,"value":13845},"        m ",{"type":21,"tag":322,"props":13847,"children":13848},{"style":3638},[13849],{"type":26,"value":3678},{"type":21,"tag":322,"props":13851,"children":13852},{"style":329},[13853],{"type":26,"value":13804},{"type":21,"tag":322,"props":13855,"children":13856},{"style":3638},[13857],{"type":26,"value":5285},{"type":21,"tag":322,"props":13859,"children":13860},{"style":348},[13861],{"type":26,"value":13231},{"type":21,"tag":322,"props":13863,"children":13864},{"class":324,"line":431},[13865],{"type":21,"tag":322,"props":13866,"children":13867},{"style":329},[13868],{"type":26,"value":13869},"        qc.h(m)\n",{"type":21,"tag":322,"props":13871,"children":13872},{"class":324,"line":453},[13873,13877,13881,13885,13889],{"type":21,"tag":322,"props":13874,"children":13875},{"style":3638},[13876],{"type":26,"value":5308},{"type":21,"tag":322,"props":13878,"children":13879},{"style":329},[13880],{"type":26,"value":12457},{"type":21,"tag":322,"props":13882,"children":13883},{"style":3638},[13884],{"type":26,"value":3941},{"type":21,"tag":322,"props":13886,"children":13887},{"style":348},[13888],{"type":26,"value":3946},{"type":21,"tag":322,"props":13890,"children":13891},{"style":329},[13892],{"type":26,"value":13893},"(m):\n",{"type":21,"tag":322,"props":13895,"children":13896},{"class":324,"line":474},[13897,13902,13906,13911,13915],{"type":21,"tag":322,"props":13898,"children":13899},{"style":329},[13900],{"type":26,"value":13901},"            k ",{"type":21,"tag":322,"props":13903,"children":13904},{"style":3638},[13905],{"type":26,"value":3678},{"type":21,"tag":322,"props":13907,"children":13908},{"style":329},[13909],{"type":26,"value":13910}," m ",{"type":21,"tag":322,"props":13912,"children":13913},{"style":3638},[13914],{"type":26,"value":5285},{"type":21,"tag":322,"props":13916,"children":13917},{"style":329},[13918],{"type":26,"value":13919}," qubit\n",{"type":21,"tag":322,"props":13921,"children":13922},{"class":324,"line":496},[13923,13927,13932,13937],{"type":21,"tag":322,"props":13924,"children":13925},{"style":3638},[13926],{"type":26,"value":5336},{"type":21,"tag":322,"props":13928,"children":13929},{"style":329},[13930],{"type":26,"value":13931}," k ",{"type":21,"tag":322,"props":13933,"children":13934},{"style":3638},[13935],{"type":26,"value":13936},"\u003C=",{"type":21,"tag":322,"props":13938,"children":13939},{"style":329},[13940],{"type":26,"value":13941}," k_max:\n",{"type":21,"tag":322,"props":13943,"children":13944},{"class":324,"line":16},[13945,13950,13954,13958,13962],{"type":21,"tag":322,"props":13946,"children":13947},{"style":329},[13948],{"type":26,"value":13949},"                qc.cp(np.pi ",{"type":21,"tag":322,"props":13951,"children":13952},{"style":3638},[13953],{"type":26,"value":174},{"type":21,"tag":322,"props":13955,"children":13956},{"style":348},[13957],{"type":26,"value":6821},{"type":21,"tag":322,"props":13959,"children":13960},{"style":3638},[13961],{"type":26,"value":8746},{"type":21,"tag":322,"props":13963,"children":13964},{"style":329},[13965],{"type":26,"value":13966}," k, qubit, m)\n",{"type":21,"tag":322,"props":13968,"children":13969},{"class":324,"line":3884},[13970,13974,13978,13982,13986,13990,13994,13998],{"type":21,"tag":322,"props":13971,"children":13972},{"style":3638},[13973],{"type":26,"value":5177},{"type":21,"tag":322,"props":13975,"children":13976},{"style":329},[13977],{"type":26,"value":3936},{"type":21,"tag":322,"props":13979,"children":13980},{"style":3638},[13981],{"type":26,"value":3941},{"type":21,"tag":322,"props":13983,"children":13984},{"style":348},[13985],{"type":26,"value":3946},{"type":21,"tag":322,"props":13987,"children":13988},{"style":329},[13989],{"type":26,"value":8106},{"type":21,"tag":322,"props":13991,"children":13992},{"style":3638},[13993],{"type":26,"value":6545},{"type":21,"tag":322,"props":13995,"children":13996},{"style":348},[13997],{"type":26,"value":6821},{"type":21,"tag":322,"props":13999,"children":14000},{"style":329},[14001],{"type":26,"value":4851},{"type":21,"tag":322,"props":14003,"children":14004},{"class":324,"line":3894},[14005,14010,14014,14018,14022,14026],{"type":21,"tag":322,"props":14006,"children":14007},{"style":329},[14008],{"type":26,"value":14009},"        qc.swap(q, n ",{"type":21,"tag":322,"props":14011,"children":14012},{"style":3638},[14013],{"type":26,"value":5285},{"type":21,"tag":322,"props":14015,"children":14016},{"style":329},[14017],{"type":26,"value":3936},{"type":21,"tag":322,"props":14019,"children":14020},{"style":3638},[14021],{"type":26,"value":5285},{"type":21,"tag":322,"props":14023,"children":14024},{"style":348},[14025],{"type":26,"value":6536},{"type":21,"tag":322,"props":14027,"children":14028},{"style":329},[14029],{"type":26,"value":3738},{"type":21,"tag":322,"props":14031,"children":14032},{"class":324,"line":1034},[14033,14037],{"type":21,"tag":322,"props":14034,"children":14035},{"style":3638},[14036],{"type":26,"value":5023},{"type":21,"tag":322,"props":14038,"children":14039},{"style":329},[14040],{"type":26,"value":8215},{"type":21,"tag":22,"props":14042,"children":14043},{},[14044],{"type":26,"value":14045},"Measured fidelity against the exact 4-qubit QFT:",{"type":21,"tag":4503,"props":14047,"children":14048},{},[14049,14070],{"type":21,"tag":4507,"props":14050,"children":14051},{},[14052],{"type":21,"tag":4511,"props":14053,"children":14054},{},[14055,14060,14065],{"type":21,"tag":4515,"props":14056,"children":14057},{},[14058],{"type":26,"value":14059},"k_max",{"type":21,"tag":4515,"props":14061,"children":14062},{},[14063],{"type":26,"value":14064},"CP gates",{"type":21,"tag":4515,"props":14066,"children":14067},{},[14068],{"type":26,"value":14069},"Fidelity",{"type":21,"tag":4529,"props":14071,"children":14072},{},[14073,14089,14105,14122],{"type":21,"tag":4511,"props":14074,"children":14075},{},[14076,14080,14084],{"type":21,"tag":4536,"props":14077,"children":14078},{},[14079],{"type":26,"value":3994},{"type":21,"tag":4536,"props":14081,"children":14082},{},[14083],{"type":26,"value":6948},{"type":21,"tag":4536,"props":14085,"children":14086},{},[14087],{"type":26,"value":14088},"0.8456",{"type":21,"tag":4511,"props":14090,"children":14091},{},[14092,14096,14100],{"type":21,"tag":4536,"props":14093,"children":14094},{},[14095],{"type":26,"value":423},{"type":21,"tag":4536,"props":14097,"children":14098},{},[14099],{"type":26,"value":7119},{"type":21,"tag":4536,"props":14101,"children":14102},{},[14103],{"type":26,"value":14104},"0.9856",{"type":21,"tag":4511,"props":14106,"children":14107},{},[14108,14112,14117],{"type":21,"tag":4536,"props":14109,"children":14110},{},[14111],{"type":26,"value":6948},{"type":21,"tag":4536,"props":14113,"children":14114},{},[14115],{"type":26,"value":14116},"6",{"type":21,"tag":4536,"props":14118,"children":14119},{},[14120],{"type":26,"value":14121},"1.0000",{"type":21,"tag":4511,"props":14123,"children":14124},{},[14125,14130,14134],{"type":21,"tag":4536,"props":14126,"children":14127},{},[14128],{"type":26,"value":14129},"exact",{"type":21,"tag":4536,"props":14131,"children":14132},{},[14133],{"type":26,"value":14116},{"type":21,"tag":4536,"props":14135,"children":14136},{},[14137],{"type":26,"value":14121},{"type":21,"tag":22,"props":14139,"children":14140},{},[14141,14143,14147],{"type":26,"value":14142},"Keeping rotations down to π\u002F4 recovers 98.6% fidelity with 17% fewer two-qubit gates. On larger registers the saving is dramatic: keeping only k ≤ O(log n) rotations is provably enough for Shor's algorithm to succeed, and on a noisy device the AQFT often outperforms the exact QFT because the errors you avoid outweigh the approximation you introduce. Always benchmark both on a ",{"type":21,"tag":29,"props":14144,"children":14145},{"href":1005},[14146],{"type":26,"value":3724},{"type":26,"value":14148}," with a realistic noise model before committing.",{"type":21,"tag":59,"props":14150,"children":14152},{"id":14151},"try-this-next",[14153],{"type":26,"value":14154},"Try this next",{"type":21,"tag":3556,"props":14156,"children":14157},{},[14158,14170,14183],{"type":21,"tag":3560,"props":14159,"children":14160},{},[14161,14163,14168],{"type":26,"value":14162},"Build ",{"type":21,"tag":79,"props":14164,"children":14166},{"className":14165},[],[14167],{"type":26,"value":13676},{"type":26,"value":14169}," and use it to implement phase estimation on a T gate. You should recover θ = 1\u002F8 exactly with 3 counting qubits.",{"type":21,"tag":3560,"props":14171,"children":14172},{},[14173,14175,14181],{"type":26,"value":14174},"Run the 4-qubit QFT through ",{"type":21,"tag":79,"props":14176,"children":14178},{"className":14177},[],[14179],{"type":26,"value":14180},"transpile(qc, backend, optimization_level=3)",{"type":26,"value":14182}," for a real device and watch the CP gates decompose into CNOTs. The depth increase is sobering.",{"type":21,"tag":3560,"props":14184,"children":14185},{},[14186,14188,14192],{"type":26,"value":14187},"Compare against a variational algorithm like ",{"type":21,"tag":29,"props":14189,"children":14190},{"href":31},[14191],{"type":26,"value":14},{"type":26,"value":14193}," to see the contrast between structured algorithms with proven speedups and heuristic NISQ ones.",{"type":21,"tag":22,"props":14195,"children":14196},{},[14197],{"type":26,"value":14198},"The takeaway to keep: the QFT is fast, exact, and useless by itself. Its power is entirely in what you wrap around it.",{"type":21,"tag":703,"props":14200,"children":14201},{},[14202],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":14204},[14205,14206,14207,14208,14209,14210,14211,14212,14213],{"id":11996,"depth":335,"text":11999},{"id":12066,"depth":335,"text":12069},{"id":12124,"depth":335,"text":12127},{"id":12203,"depth":335,"text":12206},{"id":12839,"depth":335,"text":12842},{"id":13072,"depth":335,"text":13075},{"id":13651,"depth":335,"text":13654},{"id":13721,"depth":335,"text":13724},{"id":14151,"depth":335,"text":14154},"content:blog:quantum-fourier-transform-tutorial.md","blog\u002Fquantum-fourier-transform-tutorial.md","blog\u002Fquantum-fourier-transform-tutorial",{"_path":14218,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":14219,"description":14220,"date":3500,"author":11,"tags":14221,"readingTime":16,"body":14224,"_type":715,"_id":16073,"_source":717,"_file":16074,"_stem":16075,"_extension":720},"\u002Fblog\u002Fquantum-machine-learning-reality-check","Quantum Machine Learning: A Reality Check","What QML is, what the evidence currently shows about whether it beats classical models, and which parts of it are genuinely promising: grounded in the Biamonte review, the barren plateaus result, and a 2026 head-to-head comparison.",[14222,14223,7585],"QML","PennyLane",{"type":18,"children":14225,"toc":16064},[14226,14231,14243,14248,14254,14259,14290,14300,14316,14326,14332,14344,14356,14361,14371,14377,14389,14401,14412,14417,14423,14428,14439,14444,14477,14482,14494,14500,14512,15845,15858,15913,15925,15944,15950,15955,15978,15988,15998,16004,16020,16032,16042,16060],{"type":21,"tag":22,"props":14227,"children":14228},{},[14229],{"type":26,"value":14230},"\"Quantum computing will supercharge AI\" is the single most repeated claim in quantum marketing. It is also the claim with the weakest evidence behind it.",{"type":21,"tag":22,"props":14232,"children":14233},{},[14234,14236,14241],{"type":26,"value":14235},"This isn't a takedown. Quantum machine learning is a real research area with real results and a handful of genuinely interesting open directions. But the gap between what QML is often ",{"type":21,"tag":51,"props":14237,"children":14238},{},[14239],{"type":26,"value":14240},"said",{"type":26,"value":14242}," to do and what has been demonstrated is wider than in almost any other corner of the field. If you're deciding whether to invest time here, you deserve the honest version.",{"type":21,"tag":22,"props":14244,"children":14245},{},[14246],{"type":26,"value":14247},"So: what QML is, what the evidence shows, and what's still worth watching.",{"type":21,"tag":59,"props":14249,"children":14251},{"id":14250},"what-quantum-machine-learning-means",[14252],{"type":26,"value":14253},"What quantum machine learning means",{"type":21,"tag":22,"props":14255,"children":14256},{},[14257],{"type":26,"value":14258},"The term covers several quite different things that get blurred together.",{"type":21,"tag":22,"props":14260,"children":14261},{},[14262,14267,14269,14275,14277,14282,14284,14288],{"type":21,"tag":71,"props":14263,"children":14264},{},[14265],{"type":26,"value":14266},"Variational circuits as trainable models.",{"type":26,"value":14268}," This is what most people building QML today are doing. You write a parameterized quantum circuit, a ",{"type":21,"tag":29,"props":14270,"children":14272},{"href":14271},"\u002Fglossary\u002Fvariational-circuit",[14273],{"type":26,"value":14274},"variational circuit",{"type":26,"value":14276},", feed data in, measure an output, and use a classical optimizer to adjust the parameters until the output matches your labels. Structurally it's the same ",{"type":21,"tag":29,"props":14278,"children":14279},{"href":978},[14280],{"type":26,"value":14281},"hybrid algorithm",{"type":26,"value":14283}," pattern as ",{"type":21,"tag":29,"props":14285,"children":14286},{"href":922},[14287],{"type":26,"value":883},{"type":26,"value":14289},", only with a loss function instead of an energy. The circuit is the model. The rotation angles are the weights.",{"type":21,"tag":22,"props":14291,"children":14292},{},[14293,14298],{"type":21,"tag":71,"props":14294,"children":14295},{},[14296],{"type":26,"value":14297},"Quantum kernels.",{"type":26,"value":14299}," Instead of training a quantum model directly, you use the quantum computer only to compute similarities between data points: encode two inputs into quantum states and measure their overlap. That similarity matrix then feeds an entirely classical support vector machine. The quantum device does one narrow job, which makes this approach far more resilient to noise than end-to-end variational training.",{"type":21,"tag":22,"props":14301,"children":14302},{},[14303,14308,14310,14314],{"type":21,"tag":71,"props":14304,"children":14305},{},[14306],{"type":26,"value":14307},"Data encoding.",{"type":26,"value":14309}," Both approaches need classical numbers turned into quantum states, and this choice matters more than anything else in the pipeline. Angle encoding maps each feature to a rotation angle. Simple, but it needs one ",{"type":21,"tag":29,"props":14311,"children":14312},{"href":1794},[14313],{"type":26,"value":3219},{"type":26,"value":14315}," per feature. Amplitude encoding packs 2ⁿ values into n qubits, which sounds magical until you look at the circuit depth required to prepare that state.",{"type":21,"tag":22,"props":14317,"children":14318},{},[14319,14324],{"type":21,"tag":71,"props":14320,"children":14321},{},[14322],{"type":26,"value":14323},"Quantum-accelerated linear algebra.",{"type":26,"value":14325}," The original, most theoretically ambitious branch: algorithms like HHL for solving linear systems with exponential speedups on paper. This is where the big speedup numbers in QML pitches come from, and where the caveats bite hardest.",{"type":21,"tag":59,"props":14327,"children":14329},{"id":14328},"the-data-loading-bottleneck",[14330],{"type":26,"value":14331},"The data-loading bottleneck",{"type":21,"tag":22,"props":14333,"children":14334},{},[14335,14336,14342],{"type":26,"value":11508},{"type":21,"tag":29,"props":14337,"children":14339},{"href":14338},"\u002Fresearch\u002Fbiamonte-quantum-machine-learning-2017",[14340],{"type":26,"value":14341},"Biamonte et al. review",{"type":26,"value":14343},", still the standard reference for the field, is unusually direct about the central problem, and it applies to that fourth category above.",{"type":21,"tag":22,"props":14345,"children":14346},{},[14347,14349,14354],{"type":26,"value":14348},"Many quantum ML algorithms with proven exponential speedups assume your data is ",{"type":21,"tag":51,"props":14350,"children":14351},{},[14352],{"type":26,"value":14353},"already",{"type":26,"value":14355}," sitting in quantum RAM, in a convenient superposition, ready to be operated on. That's a big assumption. If you have to load N classical data points into a quantum state, and loading takes time proportional to N, then an algorithm that runs in log(N) time once loaded still costs you O(N) overall. The speedup evaporates in the preprocessing step.",{"type":21,"tag":22,"props":14357,"children":14358},{},[14359],{"type":26,"value":14360},"This isn't a hardware engineering problem that better devices will fix. It's structural. And it got worse: over the past decade, a series of \"dequantization\" results showed that several proposed quantum ML speedups had classical algorithms achieving comparable scaling once you granted the classical side similar sampling access to the data. The quantum advantage in those cases turned out to be an artifact of comparing a quantum algorithm with quantum data access against a classical algorithm without equivalent access.",{"type":21,"tag":22,"props":14362,"children":14363},{},[14364,14366],{"type":26,"value":14365},"The practical rule: ",{"type":21,"tag":71,"props":14367,"children":14368},{},[14369],{"type":26,"value":14370},"be suspicious of any quantum ML speedup claim that doesn't account for how the data got in.",{"type":21,"tag":59,"props":14372,"children":14374},{"id":14373},"barren-plateaus",[14375],{"type":26,"value":14376},"Barren plateaus",{"type":21,"tag":22,"props":14378,"children":14379},{},[14380,14382,14388],{"type":26,"value":14381},"The variational approach (the one most people build with) has its own structural obstacle, and it's the one described in ",{"type":21,"tag":29,"props":14383,"children":14385},{"href":14384},"\u002Fresearch\u002Fmcclean-barren-plateaus-2018",[14386],{"type":26,"value":14387},"McClean et al.",{"type":26,"value":534},{"type":21,"tag":22,"props":14390,"children":14391},{},[14392,14394,14399],{"type":26,"value":14393},"Training a variational model means computing gradients and stepping downhill. McClean and colleagues showed that for randomly initialized circuits, the ",{"type":21,"tag":51,"props":14395,"children":14396},{},[14397],{"type":26,"value":14398},"variance",{"type":26,"value":14400}," of the gradient shrinks exponentially as you add qubits. The optimization terrain flattens into a featureless plain (a barren plateau) where every direction looks the same.",{"type":21,"tag":22,"props":14402,"children":14403},{},[14404,14406,14410],{"type":26,"value":14405},"Why this is fatal rather than merely annoying: gradients on quantum hardware are estimated from measurement statistics, so they come with ",{"type":21,"tag":29,"props":14407,"children":14408},{"href":6145},[14409],{"type":26,"value":6148},{"type":26,"value":14411},". Once the true gradient is smaller than your sampling error, you need exponentially many shots only to tell which way is down. At a few dozen qubits with a generic ansatz, the training signal is gone.",{"type":21,"tag":22,"props":14413,"children":14414},{},[14415],{"type":26,"value":14416},"Mitigations exist and are actively researched: structured problem-informed ansätze, layerwise training, smart initialization, local rather than global cost functions. None is a general solution. Any claim that a variational quantum model scales to useful problem sizes has to explain how it escapes this, and most don't.",{"type":21,"tag":59,"props":14418,"children":14420},{"id":14419},"the-2026-head-to-head-comparison",[14421],{"type":26,"value":14422},"The 2026 head-to-head comparison",{"type":21,"tag":22,"props":14424,"children":14425},{},[14426],{"type":26,"value":14427},"Theoretical caveats are one thing. The more useful question is what happens when someone runs the models side by side, and that's rarer than you'd expect. QML papers usually benchmark against other QML papers.",{"type":21,"tag":22,"props":14429,"children":14430},{},[14431,14437],{"type":21,"tag":29,"props":14432,"children":14434},{"href":14433},"\u002Fresearch\u002Fquantum-vs-classical-ml-2026",[14435],{"type":26,"value":14436},"Yu et al. (2026)",{"type":26,"value":14438}," did the like-for-like version: seven matched quantum\u002Fclassical model pairs, spanning both supervised learning and reinforcement learning, run on the same problems with the comparison held as fair as possible.",{"type":21,"tag":22,"props":14440,"children":14441},{},[14442],{"type":26,"value":14443},"The quantum models lost. Not on one axis: on three:",{"type":21,"tag":3556,"props":14445,"children":14446},{},[14447,14457,14467],{"type":21,"tag":3560,"props":14448,"children":14449},{},[14450,14455],{"type":21,"tag":71,"props":14451,"children":14452},{},[14453],{"type":26,"value":14454},"Prediction performance.",{"type":26,"value":14456}," Classical baselines were more accurate.",{"type":21,"tag":3560,"props":14458,"children":14459},{},[14460,14465],{"type":21,"tag":71,"props":14461,"children":14462},{},[14463],{"type":26,"value":14464},"Policy stability.",{"type":26,"value":14466}," In the reinforcement learning tasks, the quantum models produced less stable policies.",{"type":21,"tag":3560,"props":14468,"children":14469},{},[14470,14475],{"type":21,"tag":71,"props":14471,"children":14472},{},[14473],{"type":26,"value":14474},"Training time.",{"type":26,"value":14476}," The quantum models were slower to train.",{"type":21,"tag":22,"props":14478,"children":14479},{},[14480],{"type":26,"value":14481},"That last one deserves emphasis, because \"quantum is slower\" is the opposite of the entire pitch. Circuit execution overhead, the shot counts needed for gradient estimation, and the classical-quantum round trips add up.",{"type":21,"tag":22,"props":14483,"children":14484},{},[14485,14487,14492],{"type":26,"value":14486},"One careful negative result doesn't close a research field. But it's the right kind of evidence, and it points the same direction as the theory. ",{"type":21,"tag":71,"props":14488,"children":14489},{},[14490],{"type":26,"value":14491},"As of mid-2026, \"quantum will make AI faster\" is not supported by the evidence.",{"type":26,"value":14493}," If someone tells you otherwise, ask them for the head-to-head benchmark.",{"type":21,"tag":59,"props":14495,"children":14497},{"id":14496},"try-it-yourself",[14498],{"type":26,"value":14499},"Try it yourself",{"type":21,"tag":22,"props":14501,"children":14502},{},[14503,14505,14510],{"type":26,"value":14504},"The best way to develop intuition here is to build one. This is a complete variational quantum classifier in ",{"type":21,"tag":29,"props":14506,"children":14508},{"href":14507},"\u002Fsdks\u002Fpennylane",[14509],{"type":26,"value":14223},{"type":26,"value":14511},": angle encoding, a trainable entangling ansatz, and a gradient-descent loop:",{"type":21,"tag":312,"props":14513,"children":14515},{"className":3626,"code":14514,"language":3628,"meta":7,"style":7},"import pennylane as qml\nfrom pennylane import numpy as np\nfrom sklearn.datasets import make_moons\nfrom sklearn.model_selection import train_test_split\n\nn_qubits = 2\nn_layers = 3\n\ndev = qml.device(\"default.qubit\", wires=n_qubits)\n\n@qml.qnode(dev)\ndef circuit(weights, x):\n    # Data encoding: each feature becomes a rotation angle\n    qml.AngleEmbedding(x, wires=range(n_qubits), rotation=\"Y\")\n\n    # Trainable ansatz: rotations plus entangling CNOTs\n    qml.BasicEntanglerLayers(weights, wires=range(n_qubits))\n\n    return qml.expval(qml.PauliZ(0))\n\ndef variational_classifier(weights, bias, x):\n    return circuit(weights, x) + bias\n\ndef square_loss(labels, preds):\n    return np.mean((labels - qml.math.stack(preds)) ** 2)\n\ndef cost(weights, bias, X, Y):\n    preds = [variational_classifier(weights, bias, x) for x in X]\n    return square_loss(Y, preds)\n\n# Two-moons dataset, labels mapped to -1 \u002F +1 to match \u003CZ>\nX, y = make_moons(n_samples=200, noise=0.15, random_state=42)\ny = 2 * y - 1\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, test_size=0.3, random_state=42\n)\n\nshape = qml.BasicEntanglerLayers.shape(n_layers=n_layers, n_wires=n_qubits)\nrng = np.random.default_rng(42)\nweights = np.array(rng.normal(0, 0.1, shape), requires_grad=True)\nbias = np.array(0.0, requires_grad=True)\n\nopt = qml.AdamOptimizer(stepsize=0.1)\nbatch_size = 20\n\nfor step in range(60):\n    idx = np.random.randint(0, len(X_train), (batch_size,))\n    X_batch, y_batch = X_train[idx], y_train[idx]\n    weights, bias, _, _ = opt.step(cost, weights, bias, X_batch, y_batch)\n\n    if step % 15 == 0:\n        preds = [np.sign(variational_classifier(weights, bias, x)) for x in X_train]\n        acc = np.mean(np.array(preds) == y_train)\n        print(f\"Step {step:2d} | cost {cost(weights, bias, X_train, y_train):.4f} | acc {acc:.3f}\")\n\ntest_preds = [np.sign(variational_classifier(weights, bias, x)) for x in X_test]\nprint(f\"\\nTest accuracy: {np.mean(np.array(test_preds) == y_test):.3f}\")\n",[14516],{"type":21,"tag":79,"props":14517,"children":14518},{"__ignoreMap":7},[14519,14540,14567,14588,14609,14616,14632,14649,14656,14696,14703,14716,14733,14741,14784,14791,14799,14824,14831,14851,14858,14875,14896,14903,14920,14953,14960,14977,15011,15023,15030,15038,15109,15142,15159,15198,15205,15212,15256,15281,15333,15375,15382,15416,15433,15440,15473,15508,15526,15544,15552,15586,15621,15648,15739,15747,15781],{"type":21,"tag":322,"props":14520,"children":14521},{"class":324,"line":325},[14522,14526,14531,14535],{"type":21,"tag":322,"props":14523,"children":14524},{"style":3638},[14525],{"type":26,"value":3651},{"type":21,"tag":322,"props":14527,"children":14528},{"style":329},[14529],{"type":26,"value":14530}," pennylane ",{"type":21,"tag":322,"props":14532,"children":14533},{"style":3638},[14534],{"type":26,"value":4715},{"type":21,"tag":322,"props":14536,"children":14537},{"style":329},[14538],{"type":26,"value":14539}," qml\n",{"type":21,"tag":322,"props":14541,"children":14542},{"class":324,"line":335},[14543,14547,14551,14555,14559,14563],{"type":21,"tag":322,"props":14544,"children":14545},{"style":3638},[14546],{"type":26,"value":3641},{"type":21,"tag":322,"props":14548,"children":14549},{"style":329},[14550],{"type":26,"value":14530},{"type":21,"tag":322,"props":14552,"children":14553},{"style":3638},[14554],{"type":26,"value":3651},{"type":21,"tag":322,"props":14556,"children":14557},{"style":329},[14558],{"type":26,"value":4710},{"type":21,"tag":322,"props":14560,"children":14561},{"style":3638},[14562],{"type":26,"value":4715},{"type":21,"tag":322,"props":14564,"children":14565},{"style":329},[14566],{"type":26,"value":4720},{"type":21,"tag":322,"props":14568,"children":14569},{"class":324,"line":344},[14570,14574,14579,14583],{"type":21,"tag":322,"props":14571,"children":14572},{"style":3638},[14573],{"type":26,"value":3641},{"type":21,"tag":322,"props":14575,"children":14576},{"style":329},[14577],{"type":26,"value":14578}," sklearn.datasets ",{"type":21,"tag":322,"props":14580,"children":14581},{"style":3638},[14582],{"type":26,"value":3651},{"type":21,"tag":322,"props":14584,"children":14585},{"style":329},[14586],{"type":26,"value":14587}," make_moons\n",{"type":21,"tag":322,"props":14589,"children":14590},{"class":324,"line":407},[14591,14595,14600,14604],{"type":21,"tag":322,"props":14592,"children":14593},{"style":3638},[14594],{"type":26,"value":3641},{"type":21,"tag":322,"props":14596,"children":14597},{"style":329},[14598],{"type":26,"value":14599}," sklearn.model_selection ",{"type":21,"tag":322,"props":14601,"children":14602},{"style":3638},[14603],{"type":26,"value":3651},{"type":21,"tag":322,"props":14605,"children":14606},{"style":329},[14607],{"type":26,"value":14608}," train_test_split\n",{"type":21,"tag":322,"props":14610,"children":14611},{"class":324,"line":431},[14612],{"type":21,"tag":322,"props":14613,"children":14614},{"emptyLinePlaceholder":3662},[14615],{"type":26,"value":3665},{"type":21,"tag":322,"props":14617,"children":14618},{"class":324,"line":453},[14619,14624,14628],{"type":21,"tag":322,"props":14620,"children":14621},{"style":329},[14622],{"type":26,"value":14623},"n_qubits ",{"type":21,"tag":322,"props":14625,"children":14626},{"style":3638},[14627],{"type":26,"value":3678},{"type":21,"tag":322,"props":14629,"children":14630},{"style":348},[14631],{"type":26,"value":5531},{"type":21,"tag":322,"props":14633,"children":14634},{"class":324,"line":474},[14635,14640,14644],{"type":21,"tag":322,"props":14636,"children":14637},{"style":329},[14638],{"type":26,"value":14639},"n_layers ",{"type":21,"tag":322,"props":14641,"children":14642},{"style":3638},[14643],{"type":26,"value":3678},{"type":21,"tag":322,"props":14645,"children":14646},{"style":348},[14647],{"type":26,"value":14648}," 3\n",{"type":21,"tag":322,"props":14650,"children":14651},{"class":324,"line":496},[14652],{"type":21,"tag":322,"props":14653,"children":14654},{"emptyLinePlaceholder":3662},[14655],{"type":26,"value":3665},{"type":21,"tag":322,"props":14657,"children":14658},{"class":324,"line":16},[14659,14664,14668,14673,14678,14682,14687,14691],{"type":21,"tag":322,"props":14660,"children":14661},{"style":329},[14662],{"type":26,"value":14663},"dev ",{"type":21,"tag":322,"props":14665,"children":14666},{"style":3638},[14667],{"type":26,"value":3678},{"type":21,"tag":322,"props":14669,"children":14670},{"style":329},[14671],{"type":26,"value":14672}," qml.device(",{"type":21,"tag":322,"props":14674,"children":14675},{"style":359},[14676],{"type":26,"value":14677},"\"default.qubit\"",{"type":21,"tag":322,"props":14679,"children":14680},{"style":329},[14681],{"type":26,"value":367},{"type":21,"tag":322,"props":14683,"children":14684},{"style":3703},[14685],{"type":26,"value":14686},"wires",{"type":21,"tag":322,"props":14688,"children":14689},{"style":3638},[14690],{"type":26,"value":3678},{"type":21,"tag":322,"props":14692,"children":14693},{"style":329},[14694],{"type":26,"value":14695},"n_qubits)\n",{"type":21,"tag":322,"props":14697,"children":14698},{"class":324,"line":3884},[14699],{"type":21,"tag":322,"props":14700,"children":14701},{"emptyLinePlaceholder":3662},[14702],{"type":26,"value":3665},{"type":21,"tag":322,"props":14704,"children":14705},{"class":324,"line":3894},[14706,14711],{"type":21,"tag":322,"props":14707,"children":14708},{"style":4801},[14709],{"type":26,"value":14710},"@qml.qnode",{"type":21,"tag":322,"props":14712,"children":14713},{"style":329},[14714],{"type":26,"value":14715},"(dev)\n",{"type":21,"tag":322,"props":14717,"children":14718},{"class":324,"line":1034},[14719,14723,14728],{"type":21,"tag":322,"props":14720,"children":14721},{"style":3638},[14722],{"type":26,"value":4798},{"type":21,"tag":322,"props":14724,"children":14725},{"style":4801},[14726],{"type":26,"value":14727}," circuit",{"type":21,"tag":322,"props":14729,"children":14730},{"style":329},[14731],{"type":26,"value":14732},"(weights, x):\n",{"type":21,"tag":322,"props":14734,"children":14735},{"class":324,"line":3954},[14736],{"type":21,"tag":322,"props":14737,"children":14738},{"style":3888},[14739],{"type":26,"value":14740},"    # Data encoding: each feature becomes a rotation angle\n",{"type":21,"tag":322,"props":14742,"children":14743},{"class":324,"line":4002},[14744,14749,14753,14757,14761,14766,14771,14775,14780],{"type":21,"tag":322,"props":14745,"children":14746},{"style":329},[14747],{"type":26,"value":14748},"    qml.AngleEmbedding(x, ",{"type":21,"tag":322,"props":14750,"children":14751},{"style":3703},[14752],{"type":26,"value":14686},{"type":21,"tag":322,"props":14754,"children":14755},{"style":3638},[14756],{"type":26,"value":3678},{"type":21,"tag":322,"props":14758,"children":14759},{"style":348},[14760],{"type":26,"value":5379},{"type":21,"tag":322,"props":14762,"children":14763},{"style":329},[14764],{"type":26,"value":14765},"(n_qubits), ",{"type":21,"tag":322,"props":14767,"children":14768},{"style":3703},[14769],{"type":26,"value":14770},"rotation",{"type":21,"tag":322,"props":14772,"children":14773},{"style":3638},[14774],{"type":26,"value":3678},{"type":21,"tag":322,"props":14776,"children":14777},{"style":359},[14778],{"type":26,"value":14779},"\"Y\"",{"type":21,"tag":322,"props":14781,"children":14782},{"style":329},[14783],{"type":26,"value":3738},{"type":21,"tag":322,"props":14785,"children":14786},{"class":324,"line":4044},[14787],{"type":21,"tag":322,"props":14788,"children":14789},{"emptyLinePlaceholder":3662},[14790],{"type":26,"value":3665},{"type":21,"tag":322,"props":14792,"children":14793},{"class":324,"line":5068},[14794],{"type":21,"tag":322,"props":14795,"children":14796},{"style":3888},[14797],{"type":26,"value":14798},"    # Trainable ansatz: rotations plus entangling CNOTs\n",{"type":21,"tag":322,"props":14800,"children":14801},{"class":324,"line":5087},[14802,14807,14811,14815,14819],{"type":21,"tag":322,"props":14803,"children":14804},{"style":329},[14805],{"type":26,"value":14806},"    qml.BasicEntanglerLayers(weights, ",{"type":21,"tag":322,"props":14808,"children":14809},{"style":3703},[14810],{"type":26,"value":14686},{"type":21,"tag":322,"props":14812,"children":14813},{"style":3638},[14814],{"type":26,"value":3678},{"type":21,"tag":322,"props":14816,"children":14817},{"style":348},[14818],{"type":26,"value":5379},{"type":21,"tag":322,"props":14820,"children":14821},{"style":329},[14822],{"type":26,"value":14823},"(n_qubits))\n",{"type":21,"tag":322,"props":14825,"children":14826},{"class":324,"line":5095},[14827],{"type":21,"tag":322,"props":14828,"children":14829},{"emptyLinePlaceholder":3662},[14830],{"type":26,"value":3665},{"type":21,"tag":322,"props":14832,"children":14833},{"class":324,"line":5126},[14834,14838,14843,14847],{"type":21,"tag":322,"props":14835,"children":14836},{"style":3638},[14837],{"type":26,"value":5023},{"type":21,"tag":322,"props":14839,"children":14840},{"style":329},[14841],{"type":26,"value":14842}," qml.expval(qml.PauliZ(",{"type":21,"tag":322,"props":14844,"children":14845},{"style":348},[14846],{"type":26,"value":5605},{"type":21,"tag":322,"props":14848,"children":14849},{"style":329},[14850],{"type":26,"value":12806},{"type":21,"tag":322,"props":14852,"children":14853},{"class":324,"line":5171},[14854],{"type":21,"tag":322,"props":14855,"children":14856},{"emptyLinePlaceholder":3662},[14857],{"type":26,"value":3665},{"type":21,"tag":322,"props":14859,"children":14860},{"class":324,"line":5210},[14861,14865,14870],{"type":21,"tag":322,"props":14862,"children":14863},{"style":3638},[14864],{"type":26,"value":4798},{"type":21,"tag":322,"props":14866,"children":14867},{"style":4801},[14868],{"type":26,"value":14869}," variational_classifier",{"type":21,"tag":322,"props":14871,"children":14872},{"style":329},[14873],{"type":26,"value":14874},"(weights, bias, x):\n",{"type":21,"tag":322,"props":14876,"children":14877},{"class":324,"line":5228},[14878,14882,14887,14891],{"type":21,"tag":322,"props":14879,"children":14880},{"style":3638},[14881],{"type":26,"value":5023},{"type":21,"tag":322,"props":14883,"children":14884},{"style":329},[14885],{"type":26,"value":14886}," circuit(weights, x) ",{"type":21,"tag":322,"props":14888,"children":14889},{"style":3638},[14890],{"type":26,"value":6965},{"type":21,"tag":322,"props":14892,"children":14893},{"style":329},[14894],{"type":26,"value":14895}," bias\n",{"type":21,"tag":322,"props":14897,"children":14898},{"class":324,"line":5302},[14899],{"type":21,"tag":322,"props":14900,"children":14901},{"emptyLinePlaceholder":3662},[14902],{"type":26,"value":3665},{"type":21,"tag":322,"props":14904,"children":14905},{"class":324,"line":5330},[14906,14910,14915],{"type":21,"tag":322,"props":14907,"children":14908},{"style":3638},[14909],{"type":26,"value":4798},{"type":21,"tag":322,"props":14911,"children":14912},{"style":4801},[14913],{"type":26,"value":14914}," square_loss",{"type":21,"tag":322,"props":14916,"children":14917},{"style":329},[14918],{"type":26,"value":14919},"(labels, preds):\n",{"type":21,"tag":322,"props":14921,"children":14922},{"class":324,"line":5359},[14923,14927,14932,14936,14941,14945,14949],{"type":21,"tag":322,"props":14924,"children":14925},{"style":3638},[14926],{"type":26,"value":5023},{"type":21,"tag":322,"props":14928,"children":14929},{"style":329},[14930],{"type":26,"value":14931}," np.mean((labels ",{"type":21,"tag":322,"props":14933,"children":14934},{"style":3638},[14935],{"type":26,"value":5285},{"type":21,"tag":322,"props":14937,"children":14938},{"style":329},[14939],{"type":26,"value":14940}," qml.math.stack(preds)) ",{"type":21,"tag":322,"props":14942,"children":14943},{"style":3638},[14944],{"type":26,"value":5150},{"type":21,"tag":322,"props":14946,"children":14947},{"style":348},[14948],{"type":26,"value":6821},{"type":21,"tag":322,"props":14950,"children":14951},{"style":329},[14952],{"type":26,"value":3738},{"type":21,"tag":322,"props":14954,"children":14955},{"class":324,"line":5368},[14956],{"type":21,"tag":322,"props":14957,"children":14958},{"emptyLinePlaceholder":3662},[14959],{"type":26,"value":3665},{"type":21,"tag":322,"props":14961,"children":14962},{"class":324,"line":5396},[14963,14967,14972],{"type":21,"tag":322,"props":14964,"children":14965},{"style":3638},[14966],{"type":26,"value":4798},{"type":21,"tag":322,"props":14968,"children":14969},{"style":4801},[14970],{"type":26,"value":14971}," cost",{"type":21,"tag":322,"props":14973,"children":14974},{"style":329},[14975],{"type":26,"value":14976},"(weights, bias, X, Y):\n",{"type":21,"tag":322,"props":14978,"children":14979},{"class":324,"line":5427},[14980,14985,14989,14994,14998,15002,15006],{"type":21,"tag":322,"props":14981,"children":14982},{"style":329},[14983],{"type":26,"value":14984},"    preds ",{"type":21,"tag":322,"props":14986,"children":14987},{"style":3638},[14988],{"type":26,"value":3678},{"type":21,"tag":322,"props":14990,"children":14991},{"style":329},[14992],{"type":26,"value":14993}," [variational_classifier(weights, bias, x) ",{"type":21,"tag":322,"props":14995,"children":14996},{"style":3638},[14997],{"type":26,"value":3931},{"type":21,"tag":322,"props":14999,"children":15000},{"style":329},[15001],{"type":26,"value":13168},{"type":21,"tag":322,"props":15003,"children":15004},{"style":3638},[15005],{"type":26,"value":3941},{"type":21,"tag":322,"props":15007,"children":15008},{"style":329},[15009],{"type":26,"value":15010}," X]\n",{"type":21,"tag":322,"props":15012,"children":15013},{"class":324,"line":5449},[15014,15018],{"type":21,"tag":322,"props":15015,"children":15016},{"style":3638},[15017],{"type":26,"value":5023},{"type":21,"tag":322,"props":15019,"children":15020},{"style":329},[15021],{"type":26,"value":15022}," square_loss(Y, preds)\n",{"type":21,"tag":322,"props":15024,"children":15025},{"class":324,"line":5495},[15026],{"type":21,"tag":322,"props":15027,"children":15028},{"emptyLinePlaceholder":3662},[15029],{"type":26,"value":3665},{"type":21,"tag":322,"props":15031,"children":15032},{"class":324,"line":5508},[15033],{"type":21,"tag":322,"props":15034,"children":15035},{"style":3888},[15036],{"type":26,"value":15037},"# Two-moons dataset, labels mapped to -1 \u002F +1 to match \u003CZ>\n",{"type":21,"tag":322,"props":15039,"children":15040},{"class":324,"line":5516},[15041,15046,15050,15055,15060,15064,15069,15073,15078,15082,15087,15091,15096,15100,15105],{"type":21,"tag":322,"props":15042,"children":15043},{"style":329},[15044],{"type":26,"value":15045},"X, y ",{"type":21,"tag":322,"props":15047,"children":15048},{"style":3638},[15049],{"type":26,"value":3678},{"type":21,"tag":322,"props":15051,"children":15052},{"style":329},[15053],{"type":26,"value":15054}," make_moons(",{"type":21,"tag":322,"props":15056,"children":15057},{"style":3703},[15058],{"type":26,"value":15059},"n_samples",{"type":21,"tag":322,"props":15061,"children":15062},{"style":3638},[15063],{"type":26,"value":3678},{"type":21,"tag":322,"props":15065,"children":15066},{"style":348},[15067],{"type":26,"value":15068},"200",{"type":21,"tag":322,"props":15070,"children":15071},{"style":329},[15072],{"type":26,"value":367},{"type":21,"tag":322,"props":15074,"children":15075},{"style":3703},[15076],{"type":26,"value":15077},"noise",{"type":21,"tag":322,"props":15079,"children":15080},{"style":3638},[15081],{"type":26,"value":3678},{"type":21,"tag":322,"props":15083,"children":15084},{"style":348},[15085],{"type":26,"value":15086},"0.15",{"type":21,"tag":322,"props":15088,"children":15089},{"style":329},[15090],{"type":26,"value":367},{"type":21,"tag":322,"props":15092,"children":15093},{"style":3703},[15094],{"type":26,"value":15095},"random_state",{"type":21,"tag":322,"props":15097,"children":15098},{"style":3638},[15099],{"type":26,"value":3678},{"type":21,"tag":322,"props":15101,"children":15102},{"style":348},[15103],{"type":26,"value":15104},"42",{"type":21,"tag":322,"props":15106,"children":15107},{"style":329},[15108],{"type":26,"value":3738},{"type":21,"tag":322,"props":15110,"children":15111},{"class":324,"line":5534},[15112,15117,15121,15125,15129,15134,15138],{"type":21,"tag":322,"props":15113,"children":15114},{"style":329},[15115],{"type":26,"value":15116},"y ",{"type":21,"tag":322,"props":15118,"children":15119},{"style":3638},[15120],{"type":26,"value":3678},{"type":21,"tag":322,"props":15122,"children":15123},{"style":348},[15124],{"type":26,"value":6821},{"type":21,"tag":322,"props":15126,"children":15127},{"style":3638},[15128],{"type":26,"value":8782},{"type":21,"tag":322,"props":15130,"children":15131},{"style":329},[15132],{"type":26,"value":15133}," y ",{"type":21,"tag":322,"props":15135,"children":15136},{"style":3638},[15137],{"type":26,"value":5285},{"type":21,"tag":322,"props":15139,"children":15140},{"style":348},[15141],{"type":26,"value":13231},{"type":21,"tag":322,"props":15143,"children":15144},{"class":324,"line":5552},[15145,15150,15154],{"type":21,"tag":322,"props":15146,"children":15147},{"style":329},[15148],{"type":26,"value":15149},"X_train, X_test, y_train, y_test ",{"type":21,"tag":322,"props":15151,"children":15152},{"style":3638},[15153],{"type":26,"value":3678},{"type":21,"tag":322,"props":15155,"children":15156},{"style":329},[15157],{"type":26,"value":15158}," train_test_split(\n",{"type":21,"tag":322,"props":15160,"children":15161},{"class":324,"line":5560},[15162,15167,15172,15176,15181,15185,15189,15193],{"type":21,"tag":322,"props":15163,"children":15164},{"style":329},[15165],{"type":26,"value":15166},"    X, y, ",{"type":21,"tag":322,"props":15168,"children":15169},{"style":3703},[15170],{"type":26,"value":15171},"test_size",{"type":21,"tag":322,"props":15173,"children":15174},{"style":3638},[15175],{"type":26,"value":3678},{"type":21,"tag":322,"props":15177,"children":15178},{"style":348},[15179],{"type":26,"value":15180},"0.3",{"type":21,"tag":322,"props":15182,"children":15183},{"style":329},[15184],{"type":26,"value":367},{"type":21,"tag":322,"props":15186,"children":15187},{"style":3703},[15188],{"type":26,"value":15095},{"type":21,"tag":322,"props":15190,"children":15191},{"style":3638},[15192],{"type":26,"value":3678},{"type":21,"tag":322,"props":15194,"children":15195},{"style":348},[15196],{"type":26,"value":15197},"42\n",{"type":21,"tag":322,"props":15199,"children":15200},{"class":324,"line":5594},[15201],{"type":21,"tag":322,"props":15202,"children":15203},{"style":329},[15204],{"type":26,"value":3738},{"type":21,"tag":322,"props":15206,"children":15207},{"class":324,"line":5612},[15208],{"type":21,"tag":322,"props":15209,"children":15210},{"emptyLinePlaceholder":3662},[15211],{"type":26,"value":3665},{"type":21,"tag":322,"props":15213,"children":15214},{"class":324,"line":5637},[15215,15220,15224,15229,15234,15238,15243,15248,15252],{"type":21,"tag":322,"props":15216,"children":15217},{"style":329},[15218],{"type":26,"value":15219},"shape ",{"type":21,"tag":322,"props":15221,"children":15222},{"style":3638},[15223],{"type":26,"value":3678},{"type":21,"tag":322,"props":15225,"children":15226},{"style":329},[15227],{"type":26,"value":15228}," qml.BasicEntanglerLayers.shape(",{"type":21,"tag":322,"props":15230,"children":15231},{"style":3703},[15232],{"type":26,"value":15233},"n_layers",{"type":21,"tag":322,"props":15235,"children":15236},{"style":3638},[15237],{"type":26,"value":3678},{"type":21,"tag":322,"props":15239,"children":15240},{"style":329},[15241],{"type":26,"value":15242},"n_layers, ",{"type":21,"tag":322,"props":15244,"children":15245},{"style":3703},[15246],{"type":26,"value":15247},"n_wires",{"type":21,"tag":322,"props":15249,"children":15250},{"style":3638},[15251],{"type":26,"value":3678},{"type":21,"tag":322,"props":15253,"children":15254},{"style":329},[15255],{"type":26,"value":14695},{"type":21,"tag":322,"props":15257,"children":15258},{"class":324,"line":5679},[15259,15264,15268,15273,15277],{"type":21,"tag":322,"props":15260,"children":15261},{"style":329},[15262],{"type":26,"value":15263},"rng ",{"type":21,"tag":322,"props":15265,"children":15266},{"style":3638},[15267],{"type":26,"value":3678},{"type":21,"tag":322,"props":15269,"children":15270},{"style":329},[15271],{"type":26,"value":15272}," np.random.default_rng(",{"type":21,"tag":322,"props":15274,"children":15275},{"style":348},[15276],{"type":26,"value":15104},{"type":21,"tag":322,"props":15278,"children":15279},{"style":329},[15280],{"type":26,"value":3738},{"type":21,"tag":322,"props":15282,"children":15283},{"class":324,"line":5687},[15284,15289,15293,15298,15302,15306,15311,15316,15321,15325,15329],{"type":21,"tag":322,"props":15285,"children":15286},{"style":329},[15287],{"type":26,"value":15288},"weights ",{"type":21,"tag":322,"props":15290,"children":15291},{"style":3638},[15292],{"type":26,"value":3678},{"type":21,"tag":322,"props":15294,"children":15295},{"style":329},[15296],{"type":26,"value":15297}," np.array(rng.normal(",{"type":21,"tag":322,"props":15299,"children":15300},{"style":348},[15301],{"type":26,"value":5605},{"type":21,"tag":322,"props":15303,"children":15304},{"style":329},[15305],{"type":26,"value":367},{"type":21,"tag":322,"props":15307,"children":15308},{"style":348},[15309],{"type":26,"value":15310},"0.1",{"type":21,"tag":322,"props":15312,"children":15313},{"style":329},[15314],{"type":26,"value":15315},", shape), ",{"type":21,"tag":322,"props":15317,"children":15318},{"style":3703},[15319],{"type":26,"value":15320},"requires_grad",{"type":21,"tag":322,"props":15322,"children":15323},{"style":3638},[15324],{"type":26,"value":3678},{"type":21,"tag":322,"props":15326,"children":15327},{"style":348},[15328],{"type":26,"value":3715},{"type":21,"tag":322,"props":15330,"children":15331},{"style":329},[15332],{"type":26,"value":3738},{"type":21,"tag":322,"props":15334,"children":15335},{"class":324,"line":5722},[15336,15341,15345,15350,15355,15359,15363,15367,15371],{"type":21,"tag":322,"props":15337,"children":15338},{"style":329},[15339],{"type":26,"value":15340},"bias ",{"type":21,"tag":322,"props":15342,"children":15343},{"style":3638},[15344],{"type":26,"value":3678},{"type":21,"tag":322,"props":15346,"children":15347},{"style":329},[15348],{"type":26,"value":15349}," np.array(",{"type":21,"tag":322,"props":15351,"children":15352},{"style":348},[15353],{"type":26,"value":15354},"0.0",{"type":21,"tag":322,"props":15356,"children":15357},{"style":329},[15358],{"type":26,"value":367},{"type":21,"tag":322,"props":15360,"children":15361},{"style":3703},[15362],{"type":26,"value":15320},{"type":21,"tag":322,"props":15364,"children":15365},{"style":3638},[15366],{"type":26,"value":3678},{"type":21,"tag":322,"props":15368,"children":15369},{"style":348},[15370],{"type":26,"value":3715},{"type":21,"tag":322,"props":15372,"children":15373},{"style":329},[15374],{"type":26,"value":3738},{"type":21,"tag":322,"props":15376,"children":15377},{"class":324,"line":5838},[15378],{"type":21,"tag":322,"props":15379,"children":15380},{"emptyLinePlaceholder":3662},[15381],{"type":26,"value":3665},{"type":21,"tag":322,"props":15383,"children":15384},{"class":324,"line":5846},[15385,15390,15394,15399,15404,15408,15412],{"type":21,"tag":322,"props":15386,"children":15387},{"style":329},[15388],{"type":26,"value":15389},"opt ",{"type":21,"tag":322,"props":15391,"children":15392},{"style":3638},[15393],{"type":26,"value":3678},{"type":21,"tag":322,"props":15395,"children":15396},{"style":329},[15397],{"type":26,"value":15398}," qml.AdamOptimizer(",{"type":21,"tag":322,"props":15400,"children":15401},{"style":3703},[15402],{"type":26,"value":15403},"stepsize",{"type":21,"tag":322,"props":15405,"children":15406},{"style":3638},[15407],{"type":26,"value":3678},{"type":21,"tag":322,"props":15409,"children":15410},{"style":348},[15411],{"type":26,"value":15310},{"type":21,"tag":322,"props":15413,"children":15414},{"style":329},[15415],{"type":26,"value":3738},{"type":21,"tag":322,"props":15417,"children":15418},{"class":324,"line":5891},[15419,15424,15428],{"type":21,"tag":322,"props":15420,"children":15421},{"style":329},[15422],{"type":26,"value":15423},"batch_size ",{"type":21,"tag":322,"props":15425,"children":15426},{"style":3638},[15427],{"type":26,"value":3678},{"type":21,"tag":322,"props":15429,"children":15430},{"style":348},[15431],{"type":26,"value":15432}," 20\n",{"type":21,"tag":322,"props":15434,"children":15435},{"class":324,"line":5909},[15436],{"type":21,"tag":322,"props":15437,"children":15438},{"emptyLinePlaceholder":3662},[15439],{"type":26,"value":3665},{"type":21,"tag":322,"props":15441,"children":15442},{"class":324,"line":5917},[15443,15447,15452,15456,15460,15464,15469],{"type":21,"tag":322,"props":15444,"children":15445},{"style":3638},[15446],{"type":26,"value":3931},{"type":21,"tag":322,"props":15448,"children":15449},{"style":329},[15450],{"type":26,"value":15451}," step ",{"type":21,"tag":322,"props":15453,"children":15454},{"style":3638},[15455],{"type":26,"value":3941},{"type":21,"tag":322,"props":15457,"children":15458},{"style":348},[15459],{"type":26,"value":3946},{"type":21,"tag":322,"props":15461,"children":15462},{"style":329},[15463],{"type":26,"value":3758},{"type":21,"tag":322,"props":15465,"children":15466},{"style":348},[15467],{"type":26,"value":15468},"60",{"type":21,"tag":322,"props":15470,"children":15471},{"style":329},[15472],{"type":26,"value":4851},{"type":21,"tag":322,"props":15474,"children":15475},{"class":324,"line":6005},[15476,15481,15485,15490,15494,15498,15503],{"type":21,"tag":322,"props":15477,"children":15478},{"style":329},[15479],{"type":26,"value":15480},"    idx ",{"type":21,"tag":322,"props":15482,"children":15483},{"style":3638},[15484],{"type":26,"value":3678},{"type":21,"tag":322,"props":15486,"children":15487},{"style":329},[15488],{"type":26,"value":15489}," np.random.randint(",{"type":21,"tag":322,"props":15491,"children":15492},{"style":348},[15493],{"type":26,"value":5605},{"type":21,"tag":322,"props":15495,"children":15496},{"style":329},[15497],{"type":26,"value":367},{"type":21,"tag":322,"props":15499,"children":15500},{"style":348},[15501],{"type":26,"value":15502},"len",{"type":21,"tag":322,"props":15504,"children":15505},{"style":329},[15506],{"type":26,"value":15507},"(X_train), (batch_size,))\n",{"type":21,"tag":322,"props":15509,"children":15511},{"class":324,"line":15510},48,[15512,15517,15521],{"type":21,"tag":322,"props":15513,"children":15514},{"style":329},[15515],{"type":26,"value":15516},"    X_batch, y_batch ",{"type":21,"tag":322,"props":15518,"children":15519},{"style":3638},[15520],{"type":26,"value":3678},{"type":21,"tag":322,"props":15522,"children":15523},{"style":329},[15524],{"type":26,"value":15525}," X_train[idx], y_train[idx]\n",{"type":21,"tag":322,"props":15527,"children":15529},{"class":324,"line":15528},49,[15530,15535,15539],{"type":21,"tag":322,"props":15531,"children":15532},{"style":329},[15533],{"type":26,"value":15534},"    weights, bias, _, _ ",{"type":21,"tag":322,"props":15536,"children":15537},{"style":3638},[15538],{"type":26,"value":3678},{"type":21,"tag":322,"props":15540,"children":15541},{"style":329},[15542],{"type":26,"value":15543}," opt.step(cost, weights, bias, X_batch, y_batch)\n",{"type":21,"tag":322,"props":15545,"children":15547},{"class":324,"line":15546},50,[15548],{"type":21,"tag":322,"props":15549,"children":15550},{"emptyLinePlaceholder":3662},[15551],{"type":26,"value":3665},{"type":21,"tag":322,"props":15553,"children":15555},{"class":324,"line":15554},51,[15556,15560,15564,15568,15573,15578,15582],{"type":21,"tag":322,"props":15557,"children":15558},{"style":3638},[15559],{"type":26,"value":12384},{"type":21,"tag":322,"props":15561,"children":15562},{"style":329},[15563],{"type":26,"value":15451},{"type":21,"tag":322,"props":15565,"children":15566},{"style":3638},[15567],{"type":26,"value":6816},{"type":21,"tag":322,"props":15569,"children":15570},{"style":348},[15571],{"type":26,"value":15572}," 15",{"type":21,"tag":322,"props":15574,"children":15575},{"style":3638},[15576],{"type":26,"value":15577}," ==",{"type":21,"tag":322,"props":15579,"children":15580},{"style":348},[15581],{"type":26,"value":12398},{"type":21,"tag":322,"props":15583,"children":15584},{"style":329},[15585],{"type":26,"value":5356},{"type":21,"tag":322,"props":15587,"children":15589},{"class":324,"line":15588},52,[15590,15595,15599,15604,15608,15612,15616],{"type":21,"tag":322,"props":15591,"children":15592},{"style":329},[15593],{"type":26,"value":15594},"        preds ",{"type":21,"tag":322,"props":15596,"children":15597},{"style":3638},[15598],{"type":26,"value":3678},{"type":21,"tag":322,"props":15600,"children":15601},{"style":329},[15602],{"type":26,"value":15603}," [np.sign(variational_classifier(weights, bias, x)) ",{"type":21,"tag":322,"props":15605,"children":15606},{"style":3638},[15607],{"type":26,"value":3931},{"type":21,"tag":322,"props":15609,"children":15610},{"style":329},[15611],{"type":26,"value":13168},{"type":21,"tag":322,"props":15613,"children":15614},{"style":3638},[15615],{"type":26,"value":3941},{"type":21,"tag":322,"props":15617,"children":15618},{"style":329},[15619],{"type":26,"value":15620}," X_train]\n",{"type":21,"tag":322,"props":15622,"children":15624},{"class":324,"line":15623},53,[15625,15630,15634,15639,15643],{"type":21,"tag":322,"props":15626,"children":15627},{"style":329},[15628],{"type":26,"value":15629},"        acc ",{"type":21,"tag":322,"props":15631,"children":15632},{"style":3638},[15633],{"type":26,"value":3678},{"type":21,"tag":322,"props":15635,"children":15636},{"style":329},[15637],{"type":26,"value":15638}," np.mean(np.array(preds) ",{"type":21,"tag":322,"props":15640,"children":15641},{"style":3638},[15642],{"type":26,"value":5346},{"type":21,"tag":322,"props":15644,"children":15645},{"style":329},[15646],{"type":26,"value":15647}," y_train)\n",{"type":21,"tag":322,"props":15649,"children":15651},{"class":324,"line":15650},54,[15652,15656,15660,15664,15669,15673,15678,15682,15686,15691,15695,15700,15704,15708,15713,15717,15722,15727,15731,15735],{"type":21,"tag":322,"props":15653,"children":15654},{"style":348},[15655],{"type":26,"value":13330},{"type":21,"tag":322,"props":15657,"children":15658},{"style":329},[15659],{"type":26,"value":3758},{"type":21,"tag":322,"props":15661,"children":15662},{"style":3638},[15663],{"type":26,"value":3763},{"type":21,"tag":322,"props":15665,"children":15666},{"style":359},[15667],{"type":26,"value":15668},"\"Step ",{"type":21,"tag":322,"props":15670,"children":15671},{"style":348},[15672],{"type":26,"value":3773},{"type":21,"tag":322,"props":15674,"children":15675},{"style":329},[15676],{"type":26,"value":15677},"step",{"type":21,"tag":322,"props":15679,"children":15680},{"style":3638},[15681],{"type":26,"value":13395},{"type":21,"tag":322,"props":15683,"children":15684},{"style":348},[15685],{"type":26,"value":3783},{"type":21,"tag":322,"props":15687,"children":15688},{"style":359},[15689],{"type":26,"value":15690}," | cost ",{"type":21,"tag":322,"props":15692,"children":15693},{"style":348},[15694],{"type":26,"value":3773},{"type":21,"tag":322,"props":15696,"children":15697},{"style":329},[15698],{"type":26,"value":15699},"cost(weights, bias, X_train, y_train)",{"type":21,"tag":322,"props":15701,"children":15702},{"style":3638},[15703],{"type":26,"value":4085},{"type":21,"tag":322,"props":15705,"children":15706},{"style":348},[15707],{"type":26,"value":3783},{"type":21,"tag":322,"props":15709,"children":15710},{"style":359},[15711],{"type":26,"value":15712}," | acc ",{"type":21,"tag":322,"props":15714,"children":15715},{"style":348},[15716],{"type":26,"value":3773},{"type":21,"tag":322,"props":15718,"children":15719},{"style":329},[15720],{"type":26,"value":15721},"acc",{"type":21,"tag":322,"props":15723,"children":15724},{"style":3638},[15725],{"type":26,"value":15726},":.3f",{"type":21,"tag":322,"props":15728,"children":15729},{"style":348},[15730],{"type":26,"value":3783},{"type":21,"tag":322,"props":15732,"children":15733},{"style":359},[15734],{"type":26,"value":3788},{"type":21,"tag":322,"props":15736,"children":15737},{"style":329},[15738],{"type":26,"value":3738},{"type":21,"tag":322,"props":15740,"children":15742},{"class":324,"line":15741},55,[15743],{"type":21,"tag":322,"props":15744,"children":15745},{"emptyLinePlaceholder":3662},[15746],{"type":26,"value":3665},{"type":21,"tag":322,"props":15748,"children":15750},{"class":324,"line":15749},56,[15751,15756,15760,15764,15768,15772,15776],{"type":21,"tag":322,"props":15752,"children":15753},{"style":329},[15754],{"type":26,"value":15755},"test_preds ",{"type":21,"tag":322,"props":15757,"children":15758},{"style":3638},[15759],{"type":26,"value":3678},{"type":21,"tag":322,"props":15761,"children":15762},{"style":329},[15763],{"type":26,"value":15603},{"type":21,"tag":322,"props":15765,"children":15766},{"style":3638},[15767],{"type":26,"value":3931},{"type":21,"tag":322,"props":15769,"children":15770},{"style":329},[15771],{"type":26,"value":13168},{"type":21,"tag":322,"props":15773,"children":15774},{"style":3638},[15775],{"type":26,"value":3941},{"type":21,"tag":322,"props":15777,"children":15778},{"style":329},[15779],{"type":26,"value":15780}," X_test]\n",{"type":21,"tag":322,"props":15782,"children":15784},{"class":324,"line":15783},57,[15785,15789,15793,15797,15801,15806,15811,15815,15820,15824,15829,15833,15837,15841],{"type":21,"tag":322,"props":15786,"children":15787},{"style":348},[15788],{"type":26,"value":3753},{"type":21,"tag":322,"props":15790,"children":15791},{"style":329},[15792],{"type":26,"value":3758},{"type":21,"tag":322,"props":15794,"children":15795},{"style":3638},[15796],{"type":26,"value":3763},{"type":21,"tag":322,"props":15798,"children":15799},{"style":359},[15800],{"type":26,"value":3788},{"type":21,"tag":322,"props":15802,"children":15803},{"style":348},[15804],{"type":26,"value":15805},"\\n",{"type":21,"tag":322,"props":15807,"children":15808},{"style":359},[15809],{"type":26,"value":15810},"Test accuracy: ",{"type":21,"tag":322,"props":15812,"children":15813},{"style":348},[15814],{"type":26,"value":3773},{"type":21,"tag":322,"props":15816,"children":15817},{"style":329},[15818],{"type":26,"value":15819},"np.mean(np.array(test_preds) ",{"type":21,"tag":322,"props":15821,"children":15822},{"style":3638},[15823],{"type":26,"value":5346},{"type":21,"tag":322,"props":15825,"children":15826},{"style":329},[15827],{"type":26,"value":15828}," y_test)",{"type":21,"tag":322,"props":15830,"children":15831},{"style":3638},[15832],{"type":26,"value":15726},{"type":21,"tag":322,"props":15834,"children":15835},{"style":348},[15836],{"type":26,"value":3783},{"type":21,"tag":322,"props":15838,"children":15839},{"style":359},[15840],{"type":26,"value":3788},{"type":21,"tag":322,"props":15842,"children":15843},{"style":329},[15844],{"type":26,"value":3738},{"type":21,"tag":22,"props":15846,"children":15847},{},[15848,15850,15856],{"type":26,"value":15849},"Running it (PennyLane 0.45, ",{"type":21,"tag":79,"props":15851,"children":15853},{"className":15852},[],[15854],{"type":26,"value":15855},"default.qubit",{"type":26,"value":15857},") gives:",{"type":21,"tag":312,"props":15859,"children":15862},{"className":15860,"code":15861,"language":26,"meta":7,"style":7},"language-text shiki shiki-themes github-dark","Step  0 | cost 1.6158 | acc 0.336\nStep 15 | cost 0.6775 | acc 0.764\nStep 30 | cost 0.6466 | acc 0.779\nStep 45 | cost 0.6372 | acc 0.850\n\nTest accuracy: 0.817\n",[15863],{"type":21,"tag":79,"props":15864,"children":15865},{"__ignoreMap":7},[15866,15874,15882,15890,15898,15905],{"type":21,"tag":322,"props":15867,"children":15868},{"class":324,"line":325},[15869],{"type":21,"tag":322,"props":15870,"children":15871},{},[15872],{"type":26,"value":15873},"Step  0 | cost 1.6158 | acc 0.336\n",{"type":21,"tag":322,"props":15875,"children":15876},{"class":324,"line":335},[15877],{"type":21,"tag":322,"props":15878,"children":15879},{},[15880],{"type":26,"value":15881},"Step 15 | cost 0.6775 | acc 0.764\n",{"type":21,"tag":322,"props":15883,"children":15884},{"class":324,"line":344},[15885],{"type":21,"tag":322,"props":15886,"children":15887},{},[15888],{"type":26,"value":15889},"Step 30 | cost 0.6466 | acc 0.779\n",{"type":21,"tag":322,"props":15891,"children":15892},{"class":324,"line":407},[15893],{"type":21,"tag":322,"props":15894,"children":15895},{},[15896],{"type":26,"value":15897},"Step 45 | cost 0.6372 | acc 0.850\n",{"type":21,"tag":322,"props":15899,"children":15900},{"class":324,"line":431},[15901],{"type":21,"tag":322,"props":15902,"children":15903},{"emptyLinePlaceholder":3662},[15904],{"type":26,"value":3665},{"type":21,"tag":322,"props":15906,"children":15907},{"class":324,"line":453},[15908],{"type":21,"tag":322,"props":15909,"children":15910},{},[15911],{"type":26,"value":15912},"Test accuracy: 0.817\n",{"type":21,"tag":22,"props":15914,"children":15915},{},[15916,15918,15923],{"type":26,"value":15917},"So it trains, and 82% on two moons is respectable. It is also worth noting what it doesn't do: an RBF-kernel SVM from scikit-learn scores ",{"type":21,"tag":71,"props":15919,"children":15920},{},[15921],{"type":26,"value":15922},"0.983",{"type":26,"value":15924}," on the identical split, and fits in under a millisecond on your laptop. That comparison isn't a criticism of the code. It's the honest baseline that a lot of QML demos quietly omit.",{"type":21,"tag":22,"props":15926,"children":15927},{},[15928,15930,15936,15938,15942],{"type":26,"value":15929},"Try scaling it. Push ",{"type":21,"tag":79,"props":15931,"children":15933},{"className":15932},[],[15934],{"type":26,"value":15935},"n_qubits",{"type":26,"value":15937}," up, widen the ansatz, and watch the gradients shrink. You reproduce the barren-plateau effect on a ",{"type":21,"tag":29,"props":15939,"children":15940},{"href":1005},[15941],{"type":26,"value":3724},{"type":26,"value":15943}," in an afternoon, which is a far better education than reading about it.",{"type":21,"tag":59,"props":15945,"children":15947},{"id":15946},"whats-genuinely-promising",[15948],{"type":26,"value":15949},"What's genuinely promising",{"type":21,"tag":22,"props":15951,"children":15952},{},[15953],{"type":26,"value":15954},"None of the above means QML is a dead end. Three directions look genuinely worth pursuing, and they share a feature: they avoid the problems above rather than hoping to power through them.",{"type":21,"tag":22,"props":15956,"children":15957},{},[15958,15963,15965,15970,15972,15976],{"type":21,"tag":71,"props":15959,"children":15960},{},[15961],{"type":26,"value":15962},"Quantum data instead of classical data.",{"type":26,"value":15964}," The data-loading bottleneck exists because we're forcing classical numbers into quantum states. If your data is ",{"type":21,"tag":51,"props":15966,"children":15967},{},[15968],{"type":26,"value":15969},"already quantum",{"type":26,"value":15971}," (states produced by a physics experiment, molecular ground states, outputs of a quantum sensor), there's nothing to load. Learning tasks on quantum data have the cleanest theoretical case for advantage, and unlike most QML claims, the argument survives scrutiny. This overlaps heavily with where quantum computing is delivering results generally: physics and chemistry, not business analytics. Our ",{"type":21,"tag":29,"props":15973,"children":15974},{"href":11943},[15975],{"type":26,"value":11946},{"type":26,"value":15977}," covers the distinction.",{"type":21,"tag":22,"props":15979,"children":15980},{},[15981,15986],{"type":21,"tag":71,"props":15982,"children":15983},{},[15984],{"type":26,"value":15985},"Quantum kernels for structured data.",{"type":26,"value":15987}," Because the quantum device only computes similarities, kernel methods sidestep barren plateaus entirely. There's no deep variational terrain to descend. For data with structure that maps naturally onto a quantum feature space (group-theoretic structure, certain periodic problems), there are constructed examples with provable separations. The honest caveat is that these are usually engineered problems rather than datasets anyone had lying around.",{"type":21,"tag":22,"props":15989,"children":15990},{},[15991,15996],{"type":21,"tag":71,"props":15992,"children":15993},{},[15994],{"type":26,"value":15995},"Noise filtering.",{"type":26,"value":15997}," Notably, the Yu et al. comparison that found quantum models losing on every headline metric also identified noise filtering and false-positive control as areas where quantum approaches still looked promising. That's a narrow, specific niche, and narrow, specific niches are how technologies gain footholds.",{"type":21,"tag":59,"props":15999,"children":16001},{"id":16000},"where-this-leaves-you",[16002],{"type":26,"value":16003},"Where this leaves you",{"type":21,"tag":22,"props":16005,"children":16006},{},[16007,16009,16013,16014,16018],{"type":26,"value":16008},"If you're learning quantum computing, QML is still worth your time, only for the right reasons. The techniques transfer. Data encoding, ansatz design, gradient estimation, and the hybrid loop are the same skills that ",{"type":21,"tag":29,"props":16010,"children":16011},{"href":893},[16012],{"type":26,"value":883},{"type":26,"value":236},{"type":21,"tag":29,"props":16015,"children":16016},{"href":31},[16017],{"type":26,"value":14},{"type":26,"value":16019}," need, and those have clearer near-term paths. Learning QML makes you better at variational quantum computing generally.",{"type":21,"tag":22,"props":16021,"children":16022},{},[16023,16025,16030],{"type":26,"value":16024},"What to avoid is building a business case on a speedup that hasn't been demonstrated. The same discipline applies here as in ",{"type":21,"tag":29,"props":16026,"children":16027},{"href":1274},[16028],{"type":26,"value":16029},"benchmarking hardware claims",{"type":26,"value":16031},": ask what the classical baseline was, ask whether data loading was counted, ask how many qubits the result scales to.",{"type":21,"tag":22,"props":16033,"children":16034},{},[16035,16040],{"type":21,"tag":29,"props":16036,"children":16037},{"href":9190},[16038],{"type":26,"value":16039},"Preskill's NISQ paper",{"type":26,"value":16041}," set the tone for this kind of honesty about near-term devices, and it has aged well precisely because it under-promised. QML would benefit from the same posture.",{"type":21,"tag":22,"props":16043,"children":16044},{},[16045,16047,16052,16054,16058],{"type":26,"value":16046},"The field is more interesting when you stop needing it to be revolutionary. Start with the ",{"type":21,"tag":29,"props":16048,"children":16049},{"href":14507},[16050],{"type":26,"value":16051},"PennyLane SDK guide",{"type":26,"value":16053},", keep the ",{"type":21,"tag":29,"props":16055,"children":16056},{"href":4400},[16057],{"type":26,"value":9765},{"type":26,"value":16059}," open, and run the experiments yourself. The evidence is more useful than the pitch.",{"type":21,"tag":703,"props":16061,"children":16062},{},[16063],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":16065},[16066,16067,16068,16069,16070,16071,16072],{"id":14250,"depth":335,"text":14253},{"id":14328,"depth":335,"text":14331},{"id":14373,"depth":335,"text":14376},{"id":14419,"depth":335,"text":14422},{"id":14496,"depth":335,"text":14499},{"id":15946,"depth":335,"text":15949},{"id":16000,"depth":335,"text":16003},"content:blog:quantum-machine-learning-reality-check.md","blog\u002Fquantum-machine-learning-reality-check.md","blog\u002Fquantum-machine-learning-reality-check",{"_path":1457,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":16077,"description":16078,"date":3500,"author":11,"tags":16079,"readingTime":16,"body":16080,"_type":715,"_id":16454,"_source":717,"_file":16455,"_stem":16456,"_extension":720},"Quantum Networking and Distributed Quantum Computing: Wiring QPUs Together","Building one enormous quantum processor gets harder the bigger it gets. Linking smaller ones is the alternative, and it depends on distributing entanglement between machines, which is far harder than it sounds.",[730,9326,1033],{"type":18,"children":16081,"toc":16445},[16082,16087,16092,16112,16117,16123,16128,16138,16169,16179,16184,16190,16195,16206,16211,16241,16259,16265,16277,16282,16294,16299,16305,16310,16315,16320,16331,16337,16342,16351,16368,16379,16385,16390,16395,16400,16405,16411,16416,16433],{"type":21,"tag":22,"props":16083,"children":16084},{},[16085],{"type":26,"value":16086},"There is a ceiling problem in quantum hardware that doesn't get discussed as often as qubit counts do.",{"type":21,"tag":22,"props":16088,"children":16089},{},[16090],{"type":26,"value":16091},"Every approach to building a quantum processor gets harder as the device grows. Superconducting chips need a control line for essentially every qubit, and those lines carry heat into a dilution refrigerator that has a finite cooling budget. Trapped-ion chains get slower and more fragile as you add ions to a single trap. Crosstalk between neighbouring qubits rises with density. Fabrication yield (the probability that every qubit on a chip is good) falls off a cliff as area increases. None of these are fundamental physics barriers. They are engineering walls, and engineering walls have a habit of arriving sooner than expected.",{"type":21,"tag":22,"props":16093,"children":16094},{},[16095,16097,16103,16105,16110],{"type":26,"value":16096},"The alternative is the one classical computing took decades ago: stop building one bigger machine and start connecting smaller ones. That is what ",{"type":21,"tag":29,"props":16098,"children":16100},{"href":16099},"\u002Fglossary\u002Fdistributed-quantum-computing",[16101],{"type":26,"value":16102},"distributed quantum computing",{"type":26,"value":16104}," means, and it is why quantum networking has moved from a communications curiosity to a scaling strategy. When we ",{"type":21,"tag":29,"props":16106,"children":16107},{"href":1274},[16108],{"type":26,"value":16109},"read EO 14413 as a technical document",{"type":26,"value":16111},", distributed quantum computing showed up as a named five-year planning target, which is a fairly direct admission that nobody is confident the single-big-chip route gets all the way there.",{"type":21,"tag":22,"props":16113,"children":16114},{},[16115],{"type":26,"value":16116},"The catch is that connecting quantum computers is nothing like connecting classical ones.",{"type":21,"tag":59,"props":16118,"children":16120},{"id":16119},"three-different-things-people-call-quantum-networking",[16121],{"type":26,"value":16122},"Three different things people call \"quantum networking\"",{"type":21,"tag":22,"props":16124,"children":16125},{},[16126],{"type":26,"value":16127},"Before anything else, a disambiguation, because these get conflated constantly and the confusion makes most coverage of the topic unreadable.",{"type":21,"tag":22,"props":16129,"children":16130},{},[16131,16136],{"type":21,"tag":71,"props":16132,"children":16133},{},[16134],{"type":26,"value":16135},"(a) Quantum networking for distributed computation.",{"type":26,"value":16137}," Linking two or more QPUs with a quantum channel so they behave as one larger computer. Quantum information genuinely moves between them. This is the subject of this post.",{"type":21,"tag":22,"props":16139,"children":16140},{},[16141,16146,16148,16153,16155,16161,16163,16167],{"type":21,"tag":71,"props":16142,"children":16143},{},[16144],{"type":26,"value":16145},"(b) Quantum key distribution and the \"quantum internet.\"",{"type":26,"value":16147}," Using quantum states to establish a shared secret key whose security rests on physics rather than on computational hardness. This is a ",{"type":21,"tag":51,"props":16149,"children":16150},{},[16151],{"type":26,"value":16152},"communications security",{"type":26,"value":16154}," technology, descending from ",{"type":21,"tag":29,"props":16156,"children":16158},{"href":16157},"\u002Fresearch\u002Fbennett-brassard-bb84-1984",[16159],{"type":26,"value":16160},"Bennett and Brassard's BB84 protocol",{"type":26,"value":16162},". It does not make any computer faster. It's also frequently misrepresented as the answer to the quantum threat against public-key cryptography. In practice that threat is being addressed by ",{"type":21,"tag":29,"props":16164,"children":16165},{"href":9817},[16166],{"type":26,"value":9800},{"type":26,"value":16168},", which is software you deploy today, not hardware that needs new fibre.",{"type":21,"tag":22,"props":16170,"children":16171},{},[16172,16177],{"type":21,"tag":71,"props":16173,"children":16174},{},[16175],{"type":26,"value":16176},"(c) Classical networking of quantum computers.",{"type":26,"value":16178}," Running one circuit on a QPU in Maryland from a laptop in Berlin. This is what every cloud quantum service already does. It is ordinary TCP\u002FIP carrying circuit descriptions and measurement results, and nothing quantum crosses the wire.",{"type":21,"tag":22,"props":16180,"children":16181},{},[16182],{"type":26,"value":16183},"Press releases blur (a) and (c) especially often. If a headline says \"networked quantum computers\" and the mechanism is an API, it's (c).",{"type":21,"tag":59,"props":16185,"children":16187},{"id":16186},"why-you-cannot-simply-amplify-a-qubit",[16188],{"type":26,"value":16189},"Why you cannot simply amplify a qubit",{"type":21,"tag":22,"props":16191,"children":16192},{},[16193],{"type":26,"value":16194},"Classical networking works because signals are copied. A repeater reads a degraded bit, decides whether it was a 0 or a 1, and transmits a fresh clean copy. Errors are stripped out at every hop, which is why a fibre link spans an ocean.",{"type":21,"tag":22,"props":16196,"children":16197},{},[16198,16200,16204],{"type":26,"value":16199},"Quantum information forbids this. The no-cloning theorem says there is no operation that duplicates an arbitrary unknown quantum state. There is no such thing as reading a qubit to see what it is and sending a fresh one, because measurement collapses the state and destroys the ",{"type":21,"tag":29,"props":16201,"children":16202},{"href":9290},[16203],{"type":26,"value":9293},{"type":26,"value":16205}," you were trying to preserve. A quantum repeater that worked like a classical repeater would be a quantum repeater that deletes your data.",{"type":21,"tag":22,"props":16207,"children":16208},{},[16209],{"type":26,"value":16210},"Meanwhile the physical channel is brutally lossy. Photons in optical fibre attenuate exponentially with distance, and unlike classical light you can't compensate by turning up the power. A single photon either arrives or it doesn't. Beyond a few hundred kilometres, the odds of a photon surviving the trip get small enough that you'd wait a long time for one success.",{"type":21,"tag":22,"props":16212,"children":16213},{},[16214,16216,16224,16226,16231,16233,16239],{"type":26,"value":16215},"Quantum repeaters solve this by a genuinely different mechanism. Rather than relaying the qubit, they relay ",{"type":21,"tag":51,"props":16217,"children":16218},{},[16219],{"type":21,"tag":29,"props":16220,"children":16221},{"href":4286},[16222],{"type":26,"value":16223},"entanglement",{"type":26,"value":16225},". Split a long link into short segments. Establish an entangled pair across each segment independently: short hops, so the loss per hop is manageable, and success is heralded, meaning you know when it worked and simply retry when it didn't. Then perform ",{"type":21,"tag":71,"props":16227,"children":16228},{},[16229],{"type":26,"value":16230},"entanglement swapping",{"type":26,"value":16232},": at each intermediate node, jointly measure the two local halves of two adjacent pairs in the ",{"type":21,"tag":29,"props":16234,"children":16236},{"href":16235},"\u002Fglossary\u002Fbell-state",[16237],{"type":26,"value":16238},"Bell basis",{"type":26,"value":16240},". That measurement consumes both short pairs and leaves the two far-end qubits entangled with each other, despite never having interacted. Chain this along the route and you have entanglement spanning the whole distance.",{"type":21,"tag":22,"props":16242,"children":16243},{},[16244,16246,16250,16252,16257],{"type":26,"value":16245},"The ingredients this demands are exactly the hard parts: heralded entanglement generation, quantum memories that hold a half-pair coherently while the neighbouring segment keeps retrying, and some form of purification or error correction to keep ",{"type":21,"tag":29,"props":16247,"children":16248},{"href":9450},[16249],{"type":26,"value":9453},{"type":26,"value":16251}," from degrading across hops. ",{"type":21,"tag":29,"props":16253,"children":16254},{"href":9366},[16255],{"type":26,"value":16256},"Decoherence",{"type":26,"value":16258}," in the memory sets a hard clock on the whole procedure. If a stored qubit dies while its partner segment is still failing, the attempt is wasted.",{"type":21,"tag":59,"props":16260,"children":16262},{"id":16261},"teleportation-is-the-transport-primitive",[16263],{"type":26,"value":16264},"Teleportation is the transport primitive",{"type":21,"tag":22,"props":16266,"children":16267},{},[16268,16270,16275],{"type":26,"value":16269},"Once two nodes share an entangled pair, moving an actual qubit between them is a solved protocol: ",{"type":21,"tag":29,"props":16271,"children":16272},{"href":4278},[16273],{"type":26,"value":16274},"quantum teleportation",{"type":26,"value":16276},", from Bennett and colleagues in 1993.",{"type":21,"tag":22,"props":16278,"children":16279},{},[16280],{"type":26,"value":16281},"The sender performs a joint measurement on the qubit to be transmitted and their half of the entangled pair, obtaining two classical bits. Those bits are sent over an ordinary classical channel. The receiver applies one of four corrections determined by those bits, and their half of the pair becomes the original state. One entangled pair and two classical bits per qubit transported.",{"type":21,"tag":22,"props":16283,"children":16284},{},[16285,16287,16292],{"type":26,"value":16286},"Two consequences are worth stating explicitly because both get mangled in popular coverage. First, the original state is destroyed by the sender's measurement. Teleportation moves a qubit, it does not copy one, which is precisely how it stays consistent with no-cloning. Second, ",{"type":21,"tag":71,"props":16288,"children":16289},{},[16290],{"type":26,"value":16291},"the classical channel is mandatory",{"type":26,"value":16293},". Without those two bits the receiver's qubit is in a completely random state and useless. Entanglement alone signals nothing. Nothing travels faster than light, and no amount of entanglement between two labs lets them communicate without a conventional link.",{"type":21,"tag":22,"props":16295,"children":16296},{},[16297],{"type":26,"value":16298},"That second point is the whole reason quantum networks are not a physics loophole. They are a way to move fragile information, not a way to move it instantly.",{"type":21,"tag":59,"props":16300,"children":16302},{"id":16301},"the-interconnect-is-the-bottleneck",[16303],{"type":26,"value":16304},"The interconnect is the bottleneck",{"type":21,"tag":22,"props":16306,"children":16307},{},[16308],{"type":26,"value":16309},"Now the engineering reality of stitching QPUs together.",{"type":21,"tag":22,"props":16311,"children":16312},{},[16313],{"type":26,"value":16314},"Inside a single processor, two-qubit gates take microseconds or less and land fidelities that, on good hardware, exceed 99.9%. Between two processors, you must first generate a shared entangled pair over a photonic link (a probabilistic, lossy process) and only then teleport a gate or a qubit across it. Inter-node entanglement generation is currently orders of magnitude slower than local gates, and arrives at meaningfully lower fidelity.",{"type":21,"tag":22,"props":16316,"children":16317},{},[16318],{"type":26,"value":16319},"This changes how you have to think about programming such a machine. A distributed QPU is not a flat pool of qubits. It is a strongly non-uniform architecture where some pairs of qubits are cheap to entangle and others are extremely expensive, and where the expensive operations also happen to be the noisy ones. Circuit compilers have to partition algorithms to minimise cross-node operations, in much the same spirit as minimising communication in classical HPC, except the penalty for getting it wrong is not only latency but error accumulation. And because entanglement must be produced faster than the memories holding it decohere, a slow interconnect doesn't merely reduce throughput. Past a certain point it stops working at all.",{"type":21,"tag":22,"props":16321,"children":16322},{},[16323,16325,16329],{"type":26,"value":16324},"There is a genuinely appealing upside, though, and it's the reason serious people pursue this: ",{"type":21,"tag":29,"props":16326,"children":16327},{"href":751},[16328],{"type":26,"value":1632},{"type":26,"value":16330}," needs physical qubit counts in the millions for useful algorithms, and no one has a credible plan to put a million high-quality qubits in one enclosure. Modularity is one of the few routes to that number that doesn't require a single manufacturing miracle.",{"type":21,"tag":59,"props":16332,"children":16334},{"id":16333},"modality-matters-here-more-than-usual",[16335],{"type":26,"value":16336},"Modality matters here more than usual",{"type":21,"tag":22,"props":16338,"children":16339},{},[16340],{"type":26,"value":16341},"The photonic interface is where hardware platforms diverge sharply.",{"type":21,"tag":22,"props":16343,"children":16344},{},[16345,16349],{"type":21,"tag":71,"props":16346,"children":16347},{},[16348],{"type":26,"value":9595},{"type":26,"value":16350}," have a natural optical interface. An ion is excited with a laser and made to emit a single photon whose polarisation or frequency is entangled with the ion's internal state. The qubit is already coupled to light at optical wavelengths that fibre transmits well. IonQ argues that this photonic compatibility was a central reason it chose trapped ions in the first place, and that the platform is particularly suited to multi-QPU networking. That is an architectural argument from a vendor whose commercial strategy depends on modular scaling, and should be read as advocacy rather than a neutral survey, but the underlying physics of atom-photon coupling is standard, well-studied quantum optics and is not in dispute.",{"type":21,"tag":22,"props":16352,"children":16353},{},[16354,16359,16361,16366],{"type":21,"tag":71,"props":16355,"children":16356},{},[16357],{"type":26,"value":16358},"Superconducting qubits",{"type":26,"value":16360}," face a harder problem. They operate at microwave frequencies inside a millikelvin refrigerator, and microwave photons cannot travel any meaningful distance at room temperature. Thermal noise swamps them. Networking them optically requires ",{"type":21,"tag":71,"props":16362,"children":16363},{},[16364],{"type":26,"value":16365},"microwave-to-optical transduction",{"type":26,"value":16367},": coherently converting a gigahertz photon into a telecom-wavelength one without destroying the quantum state. This is an active and legitimate research field, with electro-optic, optomechanical, magneto-optic and atomic-ensemble approaches all under investigation, but efficiency and added noise remain the limiting factors. IonQ makes this contrast a talking point, characterising the conversion process as slow and not yet well matched to quantum computers. The general difficulty is widely acknowledged in the literature, though the framing is naturally sharper coming from a competitor. NIST, among others, is building optical channels for remote microwave entanglement precisely because the problem is considered tractable rather than hopeless.",{"type":21,"tag":22,"props":16369,"children":16370},{},[16371,16373,16377],{"type":26,"value":16372},"Neutral atoms sit closer to the ion story optically. Photonic qubits are their own case, since the information is already in flight. Our ",{"type":21,"tag":29,"props":16374,"children":16375},{"href":814},[16376],{"type":26,"value":9612},{"type":26,"value":16378}," covers the broader trade-offs.",{"type":21,"tag":59,"props":16380,"children":16382},{"id":16381},"where-this-stands",[16383],{"type":26,"value":16384},"Where this stands",{"type":21,"tag":22,"props":16386,"children":16387},{},[16388],{"type":26,"value":16389},"Plainly: early.",{"type":21,"tag":22,"props":16391,"children":16392},{},[16393],{"type":26,"value":16394},"The building blocks are real. Ion-photon entanglement (the first milestone, where a photon leaves a trapped ion carrying entanglement with it) has been demonstrated, and IonQ reports having done so outside an academic setting. Remote ion-ion entanglement has followed, with IonQ describing a setup that collects photons from two separate trap wells, interferes them at a shared detection hub, and leaves the two distant ions entangled. IonQ frames these as the first and second of four milestones on its own roadmap to photonic interconnects, with the remaining two being the transfer of that remote entanglement onto computation qubits via swap gates, and programmable multi-QPU entanglement using single-photon switching. In 2026 the company announced it had photonically interconnected two independent trapped-ion systems.",{"type":21,"tag":22,"props":16396,"children":16397},{},[16398],{"type":26,"value":16399},"Read those carefully. They are milestone announcements on a vendor's internal roadmap, and the public write-ups notably do not disclose entanglement rates, fidelities, or success probabilities: the numbers that would let anyone judge whether the link is fast and clean enough to compute across. IonQ itself acknowledges low entanglement rates as a current limitation. Its broader case (that networking is what carries quantum computing from research demos to commercial scale, and that modular systems are faster to build and easier to service than monolithic ones) is a coherent argument, but it is a company describing why its own architecture wins.",{"type":21,"tag":22,"props":16401,"children":16402},{},[16403],{"type":26,"value":16404},"What is not in doubt is that the demonstrated capability sits far below what distributed computing requires. Getting two qubits in different traps entangled occasionally is a real physics achievement. Sustaining entanglement generation at rates and fidelities that let two processors execute a single fault-tolerant algorithm together is a different order of problem, and nobody has done it.",{"type":21,"tag":59,"props":16406,"children":16408},{"id":16407},"the-takeaway",[16409],{"type":26,"value":16410},"The takeaway",{"type":21,"tag":22,"props":16412,"children":16413},{},[16414],{"type":26,"value":16415},"Quantum networking is not a faster internet and not a security product, though it shares physics with one. It is a proposed answer to a hardware scaling wall, built on entanglement distribution, entanglement swapping, and teleportation, all constrained by the fact that quantum information cannot be copied and therefore cannot be amplified.",{"type":21,"tag":22,"props":16417,"children":16418},{},[16419,16421,16426,16427,16431],{"type":26,"value":16420},"If you're evaluating claims in this space, three questions do most of the work. Is the link genuinely quantum, or is it an API? What is the entanglement generation ",{"type":21,"tag":51,"props":16422,"children":16423},{},[16424],{"type":26,"value":16425},"rate",{"type":26,"value":236},{"type":21,"tag":51,"props":16428,"children":16429},{},[16430],{"type":26,"value":9453},{"type":26,"value":16432}," between nodes, not whether entanglement was achieved once? And is the demonstration between two nodes in adjacent racks, or over a distance where photon loss genuinely bites?",{"type":21,"tag":22,"props":16434,"children":16435},{},[16436,16438,16443],{"type":26,"value":16437},"Answers to those tend to be absent from the announcements, which is itself informative. The ",{"type":21,"tag":29,"props":16439,"children":16441},{"href":16440},"\u002Fglossary\u002Fquantum-network",[16442],{"type":26,"value":9765},{"type":26,"value":16444}," covers the vocabulary if you want to read the primary sources yourself.",{"title":7,"searchDepth":335,"depth":335,"links":16446},[16447,16448,16449,16450,16451,16452,16453],{"id":16119,"depth":335,"text":16122},{"id":16186,"depth":335,"text":16189},{"id":16261,"depth":335,"text":16264},{"id":16301,"depth":335,"text":16304},{"id":16333,"depth":335,"text":16336},{"id":16381,"depth":335,"text":16384},{"id":16407,"depth":335,"text":16410},"content:blog:quantum-networking-distributed-computing.md","blog\u002Fquantum-networking-distributed-computing.md","blog\u002Fquantum-networking-distributed-computing",{"_path":16458,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":16459,"description":16460,"date":3500,"author":11,"tags":16461,"readingTime":3894,"body":16462,"_type":715,"_id":18556,"_source":717,"_file":18557,"_stem":18558,"_extension":720},"\u002Fblog\u002Fquantum-phase-estimation-tutorial","Quantum Phase Estimation: The Missing Link Between the QFT and Shor's Algorithm","A hands-on Qiskit tutorial on quantum phase estimation: phase kickback, the inverse QFT, and precision limits, with complete runnable code and real measured output.",[7585,1962,3302],{"type":18,"children":16463,"toc":18544},[16464,16482,16500,16505,16513,16519,16524,16532,16537,16548,16587,16592,16604,16610,16615,16626,16634,16658,16666,16676,16681,16689,16709,16715,16726,16731,16741,16746,16774,16778,16814,17764,17768,17776,17786,17794,17799,17806,17838,17851,17859,17879,17884,17890,17895,18139,18143,18151,18163,18175,18183,18195,18207,18213,18236,18253,18281,18379,18403,18409,18420,18431,18473,18478,18483,18540],{"type":21,"tag":22,"props":16465,"children":16466},{},[16467,16469,16474,16476,16480],{"type":26,"value":16468},"Our ",{"type":21,"tag":29,"props":16470,"children":16471},{"href":11963},[16472],{"type":26,"value":16473},"QFT tutorial",{"type":26,"value":16475}," ended on a deliberately unsatisfying note: the ",{"type":21,"tag":29,"props":16477,"children":16478},{"href":11975},[16479],{"type":26,"value":11978},{"type":26,"value":16481}," is fast, exact, and useless on its own. You can't load its input efficiently and you can't read its output. Measure a transformed basis state and you get a uniformly random bitstring.",{"type":21,"tag":22,"props":16483,"children":16484},{},[16485,16487,16491,16493,16498],{"type":26,"value":16486},"This post is the sequel that resolves that. Quantum phase estimation (QPE) is the machine that turns the ",{"type":21,"tag":51,"props":16488,"children":16489},{},[16490],{"type":26,"value":7416},{"type":26,"value":16492}," QFT into a readout primitive. It's the reason the QFT matters, and it is the algorithm sitting underneath ",{"type":21,"tag":29,"props":16494,"children":16495},{"href":10066},[16496],{"type":26,"value":16497},"Shor's",{"type":26,"value":16499},": Shor's is QPE applied to modular exponentiation, plus number theory.",{"type":21,"tag":22,"props":16501,"children":16502},{},[16503],{"type":26,"value":16504},"Read the QFT post first if you haven't. This one assumes the construction and won't rebuild it.",{"type":21,"tag":22,"props":16506,"children":16507},{},[16508],{"type":21,"tag":71,"props":16509,"children":16510},{},[16511],{"type":26,"value":16512},"All code below was executed against Qiskit 2.5.0 and Qiskit Aer 0.17.2 on Python 3.11. The outputs shown are real.",{"type":21,"tag":59,"props":16514,"children":16516},{"id":16515},"the-problem-qpe-solves",[16517],{"type":26,"value":16518},"The problem QPE solves",{"type":21,"tag":22,"props":16520,"children":16521},{},[16522],{"type":26,"value":16523},"Given a unitary U and an eigenstate |ψ⟩ satisfying",{"type":21,"tag":312,"props":16525,"children":16527},{"code":16526},"U|ψ⟩ = e^{2πiθ} |ψ⟩\n",[16528],{"type":21,"tag":79,"props":16529,"children":16530},{"__ignoreMap":7},[16531],{"type":26,"value":16526},{"type":21,"tag":22,"props":16533,"children":16534},{},[16535],{"type":26,"value":16536},"estimate θ ∈ [0, 1).",{"type":21,"tag":22,"props":16538,"children":16539},{},[16540,16542,16547],{"type":26,"value":16541},"That sounds like a linear-algebra exercise, not an application. It is in fact one of the most consequential problems in the field, because ",{"type":21,"tag":51,"props":16543,"children":16544},{},[16545],{"type":26,"value":16546},"enormous numbers of questions are secretly eigenvalue questions",{"type":26,"value":12855},{"type":21,"tag":3556,"props":16549,"children":16550},{},[16551,16561,16571],{"type":21,"tag":3560,"props":16552,"children":16553},{},[16554,16559],{"type":21,"tag":71,"props":16555,"children":16556},{},[16557],{"type":26,"value":16558},"What is a molecule's ground-state energy?",{"type":26,"value":16560}," Build U = e^{-iHt} for the molecular Hamiltonian H. Its eigenphases are the energy levels. Estimating θ estimates the energy.",{"type":21,"tag":3560,"props":16562,"children":16563},{},[16564,16569],{"type":21,"tag":71,"props":16565,"children":16566},{},[16567],{"type":26,"value":16568},"What is the order of a mod N?",{"type":26,"value":16570}," Build U|y⟩ = |ay mod N⟩. Its eigenphases are multiples of 1\u002Fr where r is the order. Estimating θ gives you r, and factoring falls out.",{"type":21,"tag":3560,"props":16572,"children":16573},{},[16574,16579,16581,16586],{"type":21,"tag":71,"props":16575,"children":16576},{},[16577],{"type":26,"value":16578},"How many solutions does an oracle mark?",{"type":26,"value":16580}," Amplitude estimation is QPE on the Grover operator, whose eigenphase encodes the count. That's the quantum-counting subroutine we flagged as a missing piece in the ",{"type":21,"tag":29,"props":16582,"children":16583},{"href":7581},[16584],{"type":26,"value":16585},"Grover tutorial",{"type":26,"value":534},{"type":21,"tag":22,"props":16588,"children":16589},{},[16590],{"type":26,"value":16591},"One primitive, three flagship algorithms. That's why \"estimate an eigenphase\" is worth a dedicated circuit.",{"type":21,"tag":22,"props":16593,"children":16594},{},[16595,16597,16602],{"type":26,"value":16596},"Note the precondition, which people gloss over: you need to ",{"type":21,"tag":51,"props":16598,"children":16599},{},[16600],{"type":26,"value":16601},"have",{"type":26,"value":16603}," the eigenstate |ψ⟩, or at least a state with decent overlap with it. Preparing good eigenstates is its own hard problem, and in chemistry it's often the binding constraint.",{"type":21,"tag":59,"props":16605,"children":16607},{"id":16606},"phase-kickback-the-key-insight",[16608],{"type":26,"value":16609},"Phase kickback: the key insight",{"type":21,"tag":22,"props":16611,"children":16612},{},[16613],{"type":26,"value":16614},"Here's the thing that makes QPE work, and it's worth slowing down for.",{"type":21,"tag":22,"props":16616,"children":16617},{},[16618,16620,16624],{"type":26,"value":16619},"A controlled-U applied to a control ",{"type":21,"tag":29,"props":16621,"children":16622},{"href":1794},[16623],{"type":26,"value":3219},{"type":26,"value":16625}," in state |1⟩ and a target in the eigenstate |ψ⟩ gives:",{"type":21,"tag":312,"props":16627,"children":16629},{"code":16628},"CU |1⟩|ψ⟩ = |1⟩ ⊗ U|ψ⟩ = |1⟩ ⊗ e^{2πiθ}|ψ⟩ = e^{2πiθ} |1⟩|ψ⟩\n",[16630],{"type":21,"tag":79,"props":16631,"children":16632},{"__ignoreMap":7},[16633],{"type":26,"value":16628},{"type":21,"tag":22,"props":16635,"children":16636},{},[16637,16639,16644,16646,16650,16652,16656],{"type":26,"value":16638},"The eigenstate came out unchanged. The phase e^{2πiθ} is a scalar, so it's now attached to the whole state, and since it only appeared for the |1⟩ branch of the control, it's a ",{"type":21,"tag":51,"props":16640,"children":16641},{},[16642],{"type":26,"value":16643},"relative",{"type":26,"value":16645}," phase on the control qubit. Put the control in ",{"type":21,"tag":29,"props":16647,"children":16648},{"href":9290},[16649],{"type":26,"value":9293},{"type":26,"value":16651}," with a ",{"type":21,"tag":29,"props":16653,"children":16654},{"href":7692},[16655],{"type":26,"value":12144},{"type":26,"value":16657}," and you get:",{"type":21,"tag":312,"props":16659,"children":16661},{"code":16660},"CU (|0⟩+|1⟩)\u002F√2 ⊗ |ψ⟩ = (|0⟩ + e^{2πiθ}|1⟩)\u002F√2 ⊗ |ψ⟩\n",[16662],{"type":21,"tag":79,"props":16663,"children":16664},{"__ignoreMap":7},[16665],{"type":26,"value":16660},{"type":21,"tag":22,"props":16667,"children":16668},{},[16669,16674],{"type":21,"tag":71,"props":16670,"children":16671},{},[16672],{"type":26,"value":16673},"The phase kicked back from the target onto the control.",{"type":26,"value":16675}," The target register is untouched and unentangled. It's a catalyst. All of the information now lives in the control's phase.",{"type":21,"tag":22,"props":16677,"children":16678},{},[16679],{"type":26,"value":16680},"Do this with n control qubits, and give control qubit k a controlled-U^(2^k) instead, so it picks up phase e^{2πi·2^k·θ}. The counting register ends up in:",{"type":21,"tag":312,"props":16682,"children":16684},{"code":16683},"(1\u002F√N) Σ_{x=0}^{N-1} e^{2πi·x·θ} |x⟩\n",[16685],{"type":21,"tag":79,"props":16686,"children":16687},{"__ignoreMap":7},[16688],{"type":26,"value":16683},{"type":21,"tag":22,"props":16690,"children":16691},{},[16692,16694,16700,16702,16707],{"type":26,"value":16693},"Stare at that and compare it to the QFT formula from the previous post: ",{"type":21,"tag":79,"props":16695,"children":16697},{"className":16696},[],[16698],{"type":26,"value":16699},"QFT|j⟩ = (1\u002F√N) Σ_x e^{2πi·jx\u002FN}|x⟩",{"type":26,"value":16701},". They are the ",{"type":21,"tag":51,"props":16703,"children":16704},{},[16705],{"type":26,"value":16706},"same state",{"type":26,"value":16708},", with j\u002FN = θ. The counting register is holding the QFT of the number j = Nθ.",{"type":21,"tag":59,"props":16710,"children":16712},{"id":16711},"why-the-inverse-qft",[16713],{"type":26,"value":16714},"Why the inverse QFT",{"type":21,"tag":22,"props":16716,"children":16717},{},[16718,16720,16724],{"type":26,"value":16719},"So the controlled-U powers have written θ into the counting register's phases, and, as the QFT post hammered, phases are invisible to ",{"type":21,"tag":29,"props":16721,"children":16722},{"href":7857},[16723],{"type":26,"value":11538},{"type":26,"value":16725},". Measure now and you get uniform noise.",{"type":21,"tag":22,"props":16727,"children":16728},{},[16729],{"type":26,"value":16730},"But we know exactly what transform produced this state. So we undo it. Apply QFT⁻¹ and the register collapses to |j⟩ where j = Nθ, a plain binary integer you read off in one shot.",{"type":21,"tag":22,"props":16732,"children":16733},{},[16734,16736,16740],{"type":26,"value":16735},"That's the payoff the QFT post set up. The QFT direction moves position information into phases. The inverse direction pulls phase information back out into a readable number. QPE is the canonical example of the pattern we described there: the QFT is never the whole algorithm, it's the readout stage of a larger ",{"type":21,"tag":29,"props":16737,"children":16738},{"href":8484},[16739],{"type":26,"value":8487},{"type":26,"value":534},{"type":21,"tag":22,"props":16742,"children":16743},{},[16744],{"type":26,"value":16745},"The full recipe:",{"type":21,"tag":7444,"props":16747,"children":16748},{},[16749,16754,16759,16764,16769],{"type":21,"tag":3560,"props":16750,"children":16751},{},[16752],{"type":26,"value":16753},"n counting qubits + enough qubits to hold |ψ⟩.",{"type":21,"tag":3560,"props":16755,"children":16756},{},[16757],{"type":26,"value":16758},"Hadamard the counting register.",{"type":21,"tag":3560,"props":16760,"children":16761},{},[16762],{"type":26,"value":16763},"For k = 0…n−1: apply controlled-U^(2^k) from counting qubit k onto the eigenstate register.",{"type":21,"tag":3560,"props":16765,"children":16766},{},[16767],{"type":26,"value":16768},"Apply the inverse QFT to the counting register.",{"type":21,"tag":3560,"props":16770,"children":16771},{},[16772],{"type":26,"value":16773},"Measure the counting register. The result, read as a binary fraction, is θ.",{"type":21,"tag":59,"props":16775,"children":16776},{"id":12203},[16777],{"type":26,"value":12206},{"type":21,"tag":22,"props":16779,"children":16780},{},[16781,16783,16789,16791,16797,16799,16804,16806,16812],{"type":26,"value":16782},"Simplest possible U with a known answer: the phase gate ",{"type":21,"tag":79,"props":16784,"children":16786},{"className":16785},[],[16787],{"type":26,"value":16788},"P(λ)",{"type":26,"value":16790},", which maps |1⟩ → e^{iλ}|1⟩. So |1⟩ is an eigenstate, and setting λ = 2πθ makes the eigenphase exactly θ. We'll reuse the ",{"type":21,"tag":79,"props":16792,"children":16794},{"className":16793},[],[16795],{"type":26,"value":16796},"qft",{"type":26,"value":16798}," function from the ",{"type":21,"tag":29,"props":16800,"children":16801},{"href":11963},[16802],{"type":26,"value":16803},"previous post",{"type":26,"value":16805}," and simply call ",{"type":21,"tag":79,"props":16807,"children":16809},{"className":16808},[],[16810],{"type":26,"value":16811},".inverse()",{"type":26,"value":16813}," on it.",{"type":21,"tag":312,"props":16815,"children":16817},{"code":16816,"language":3628,"meta":7,"className":3626,"style":7},"import numpy as np\nfrom qiskit import QuantumCircuit, transpile\nfrom qiskit_aer import AerSimulator\n\n\ndef qft_rotations(circuit, n):\n    if n == 0:\n        return circuit\n    n -= 1\n    circuit.h(n)\n    for qubit in range(n):\n        circuit.cp(np.pi \u002F 2 ** (n - qubit), qubit, n)\n    qft_rotations(circuit, n)\n    return circuit\n\n\ndef qft(n):\n    qc = QuantumCircuit(n, name=\"QFT\")\n    qft_rotations(qc, n)\n    for q in range(n \u002F\u002F 2):          # the swaps matter here, see below\n        qc.swap(q, n - q - 1)\n    return qc\n\n\ndef qpe_phase_gate(n_count, theta):\n    \"\"\"QPE for U = P(2*pi*theta), whose eigenstate is |1>.\"\"\"\n    qc = QuantumCircuit(n_count + 1, n_count)\n    qc.h(range(n_count))             # counting register in superposition\n    qc.x(n_count)                    # prepare the eigenstate |1>\n\n    for k in range(n_count):         # controlled-U^(2^k) = CP(2*pi*theta*2^k)\n        qc.cp(2 * np.pi * theta * 2 ** k, k, n_count)\n\n    qc.compose(qft(n_count).inverse(), qubits=range(n_count), inplace=True)\n    qc.measure(range(n_count), range(n_count))\n    return qc\n\n\nsim = AerSimulator()\nqc = qpe_phase_gate(4, 0.625)        # theta = 0.625 = 0.1010 in binary\ncounts = sim.run(transpile(qc, sim), shots=4096).result().get_counts()\n\nfor bits, c in sorted(counts.items(), key=lambda kv: -kv[1]):\n    est = int(bits, 2) \u002F 2 ** 4\n    print(f\"{bits} -> theta = {int(bits, 2)}\u002F16 = {est:.4f}   counts = {c}\")\n",[16818],{"type":21,"tag":79,"props":16819,"children":16820},{"__ignoreMap":7},[16821,16840,16859,16878,16885,16892,16907,16930,16941,16956,16963,16986,17017,17025,17036,17043,17050,17065,17096,17103,17144,17171,17182,17189,17196,17213,17221,17250,17271,17284,17291,17320,17367,17374,17415,17439,17450,17457,17464,17479,17518,17549,17556,17610,17656],{"type":21,"tag":322,"props":16822,"children":16823},{"class":324,"line":325},[16824,16828,16832,16836],{"type":21,"tag":322,"props":16825,"children":16826},{"style":3638},[16827],{"type":26,"value":3651},{"type":21,"tag":322,"props":16829,"children":16830},{"style":329},[16831],{"type":26,"value":4710},{"type":21,"tag":322,"props":16833,"children":16834},{"style":3638},[16835],{"type":26,"value":4715},{"type":21,"tag":322,"props":16837,"children":16838},{"style":329},[16839],{"type":26,"value":4720},{"type":21,"tag":322,"props":16841,"children":16842},{"class":324,"line":335},[16843,16847,16851,16855],{"type":21,"tag":322,"props":16844,"children":16845},{"style":3638},[16846],{"type":26,"value":3641},{"type":21,"tag":322,"props":16848,"children":16849},{"style":329},[16850],{"type":26,"value":4732},{"type":21,"tag":322,"props":16852,"children":16853},{"style":3638},[16854],{"type":26,"value":3651},{"type":21,"tag":322,"props":16856,"children":16857},{"style":329},[16858],{"type":26,"value":4741},{"type":21,"tag":322,"props":16860,"children":16861},{"class":324,"line":344},[16862,16866,16870,16874],{"type":21,"tag":322,"props":16863,"children":16864},{"style":3638},[16865],{"type":26,"value":3641},{"type":21,"tag":322,"props":16867,"children":16868},{"style":329},[16869],{"type":26,"value":4753},{"type":21,"tag":322,"props":16871,"children":16872},{"style":3638},[16873],{"type":26,"value":3651},{"type":21,"tag":322,"props":16875,"children":16876},{"style":329},[16877],{"type":26,"value":4762},{"type":21,"tag":322,"props":16879,"children":16880},{"class":324,"line":407},[16881],{"type":21,"tag":322,"props":16882,"children":16883},{"emptyLinePlaceholder":3662},[16884],{"type":26,"value":3665},{"type":21,"tag":322,"props":16886,"children":16887},{"class":324,"line":431},[16888],{"type":21,"tag":322,"props":16889,"children":16890},{"emptyLinePlaceholder":3662},[16891],{"type":26,"value":3665},{"type":21,"tag":322,"props":16893,"children":16894},{"class":324,"line":453},[16895,16899,16903],{"type":21,"tag":322,"props":16896,"children":16897},{"style":3638},[16898],{"type":26,"value":4798},{"type":21,"tag":322,"props":16900,"children":16901},{"style":4801},[16902],{"type":26,"value":12363},{"type":21,"tag":322,"props":16904,"children":16905},{"style":329},[16906],{"type":26,"value":12368},{"type":21,"tag":322,"props":16908,"children":16909},{"class":324,"line":474},[16910,16914,16918,16922,16926],{"type":21,"tag":322,"props":16911,"children":16912},{"style":3638},[16913],{"type":26,"value":12384},{"type":21,"tag":322,"props":16915,"children":16916},{"style":329},[16917],{"type":26,"value":12389},{"type":21,"tag":322,"props":16919,"children":16920},{"style":3638},[16921],{"type":26,"value":5346},{"type":21,"tag":322,"props":16923,"children":16924},{"style":348},[16925],{"type":26,"value":12398},{"type":21,"tag":322,"props":16927,"children":16928},{"style":329},[16929],{"type":26,"value":5356},{"type":21,"tag":322,"props":16931,"children":16932},{"class":324,"line":496},[16933,16937],{"type":21,"tag":322,"props":16934,"children":16935},{"style":3638},[16936],{"type":26,"value":12410},{"type":21,"tag":322,"props":16938,"children":16939},{"style":329},[16940],{"type":26,"value":12415},{"type":21,"tag":322,"props":16942,"children":16943},{"class":324,"line":16},[16944,16948,16952],{"type":21,"tag":322,"props":16945,"children":16946},{"style":329},[16947],{"type":26,"value":12423},{"type":21,"tag":322,"props":16949,"children":16950},{"style":3638},[16951],{"type":26,"value":12428},{"type":21,"tag":322,"props":16953,"children":16954},{"style":348},[16955],{"type":26,"value":13231},{"type":21,"tag":322,"props":16957,"children":16958},{"class":324,"line":3884},[16959],{"type":21,"tag":322,"props":16960,"children":16961},{"style":329},[16962],{"type":26,"value":12445},{"type":21,"tag":322,"props":16964,"children":16965},{"class":324,"line":3894},[16966,16970,16974,16978,16982],{"type":21,"tag":322,"props":16967,"children":16968},{"style":3638},[16969],{"type":26,"value":5177},{"type":21,"tag":322,"props":16971,"children":16972},{"style":329},[16973],{"type":26,"value":12457},{"type":21,"tag":322,"props":16975,"children":16976},{"style":3638},[16977],{"type":26,"value":3941},{"type":21,"tag":322,"props":16979,"children":16980},{"style":348},[16981],{"type":26,"value":3946},{"type":21,"tag":322,"props":16983,"children":16984},{"style":329},[16985],{"type":26,"value":8278},{"type":21,"tag":322,"props":16987,"children":16988},{"class":324,"line":1034},[16989,16993,16997,17001,17005,17009,17013],{"type":21,"tag":322,"props":16990,"children":16991},{"style":329},[16992],{"type":26,"value":12485},{"type":21,"tag":322,"props":16994,"children":16995},{"style":3638},[16996],{"type":26,"value":174},{"type":21,"tag":322,"props":16998,"children":16999},{"style":348},[17000],{"type":26,"value":6821},{"type":21,"tag":322,"props":17002,"children":17003},{"style":3638},[17004],{"type":26,"value":8746},{"type":21,"tag":322,"props":17006,"children":17007},{"style":329},[17008],{"type":26,"value":12502},{"type":21,"tag":322,"props":17010,"children":17011},{"style":3638},[17012],{"type":26,"value":5285},{"type":21,"tag":322,"props":17014,"children":17015},{"style":329},[17016],{"type":26,"value":12511},{"type":21,"tag":322,"props":17018,"children":17019},{"class":324,"line":3954},[17020],{"type":21,"tag":322,"props":17021,"children":17022},{"style":329},[17023],{"type":26,"value":17024},"    qft_rotations(circuit, n)\n",{"type":21,"tag":322,"props":17026,"children":17027},{"class":324,"line":4002},[17028,17032],{"type":21,"tag":322,"props":17029,"children":17030},{"style":3638},[17031],{"type":26,"value":5023},{"type":21,"tag":322,"props":17033,"children":17034},{"style":329},[17035],{"type":26,"value":12415},{"type":21,"tag":322,"props":17037,"children":17038},{"class":324,"line":4044},[17039],{"type":21,"tag":322,"props":17040,"children":17041},{"emptyLinePlaceholder":3662},[17042],{"type":26,"value":3665},{"type":21,"tag":322,"props":17044,"children":17045},{"class":324,"line":5068},[17046],{"type":21,"tag":322,"props":17047,"children":17048},{"emptyLinePlaceholder":3662},[17049],{"type":26,"value":3665},{"type":21,"tag":322,"props":17051,"children":17052},{"class":324,"line":5087},[17053,17057,17061],{"type":21,"tag":322,"props":17054,"children":17055},{"style":3638},[17056],{"type":26,"value":4798},{"type":21,"tag":322,"props":17058,"children":17059},{"style":4801},[17060],{"type":26,"value":12673},{"type":21,"tag":322,"props":17062,"children":17063},{"style":329},[17064],{"type":26,"value":8278},{"type":21,"tag":322,"props":17066,"children":17067},{"class":324,"line":5095},[17068,17072,17076,17080,17084,17088,17092],{"type":21,"tag":322,"props":17069,"children":17070},{"style":329},[17071],{"type":26,"value":7934},{"type":21,"tag":322,"props":17073,"children":17074},{"style":3638},[17075],{"type":26,"value":3678},{"type":21,"tag":322,"props":17077,"children":17078},{"style":329},[17079],{"type":26,"value":7943},{"type":21,"tag":322,"props":17081,"children":17082},{"style":3703},[17083],{"type":26,"value":7948},{"type":21,"tag":322,"props":17085,"children":17086},{"style":3638},[17087],{"type":26,"value":3678},{"type":21,"tag":322,"props":17089,"children":17090},{"style":359},[17091],{"type":26,"value":12705},{"type":21,"tag":322,"props":17093,"children":17094},{"style":329},[17095],{"type":26,"value":3738},{"type":21,"tag":322,"props":17097,"children":17098},{"class":324,"line":5126},[17099],{"type":21,"tag":322,"props":17100,"children":17101},{"style":329},[17102],{"type":26,"value":12717},{"type":21,"tag":322,"props":17104,"children":17105},{"class":324,"line":5171},[17106,17110,17114,17118,17122,17126,17130,17134,17139],{"type":21,"tag":322,"props":17107,"children":17108},{"style":3638},[17109],{"type":26,"value":5177},{"type":21,"tag":322,"props":17111,"children":17112},{"style":329},[17113],{"type":26,"value":3936},{"type":21,"tag":322,"props":17115,"children":17116},{"style":3638},[17117],{"type":26,"value":3941},{"type":21,"tag":322,"props":17119,"children":17120},{"style":348},[17121],{"type":26,"value":3946},{"type":21,"tag":322,"props":17123,"children":17124},{"style":329},[17125],{"type":26,"value":8106},{"type":21,"tag":322,"props":17127,"children":17128},{"style":3638},[17129],{"type":26,"value":6545},{"type":21,"tag":322,"props":17131,"children":17132},{"style":348},[17133],{"type":26,"value":6821},{"type":21,"tag":322,"props":17135,"children":17136},{"style":329},[17137],{"type":26,"value":17138},"):          ",{"type":21,"tag":322,"props":17140,"children":17141},{"style":3888},[17142],{"type":26,"value":17143},"# the swaps matter here, see below\n",{"type":21,"tag":322,"props":17145,"children":17146},{"class":324,"line":5210},[17147,17151,17155,17159,17163,17167],{"type":21,"tag":322,"props":17148,"children":17149},{"style":329},[17150],{"type":26,"value":14009},{"type":21,"tag":322,"props":17152,"children":17153},{"style":3638},[17154],{"type":26,"value":5285},{"type":21,"tag":322,"props":17156,"children":17157},{"style":329},[17158],{"type":26,"value":3936},{"type":21,"tag":322,"props":17160,"children":17161},{"style":3638},[17162],{"type":26,"value":5285},{"type":21,"tag":322,"props":17164,"children":17165},{"style":348},[17166],{"type":26,"value":6536},{"type":21,"tag":322,"props":17168,"children":17169},{"style":329},[17170],{"type":26,"value":3738},{"type":21,"tag":322,"props":17172,"children":17173},{"class":324,"line":5228},[17174,17178],{"type":21,"tag":322,"props":17175,"children":17176},{"style":3638},[17177],{"type":26,"value":5023},{"type":21,"tag":322,"props":17179,"children":17180},{"style":329},[17181],{"type":26,"value":8215},{"type":21,"tag":322,"props":17183,"children":17184},{"class":324,"line":5302},[17185],{"type":21,"tag":322,"props":17186,"children":17187},{"emptyLinePlaceholder":3662},[17188],{"type":26,"value":3665},{"type":21,"tag":322,"props":17190,"children":17191},{"class":324,"line":5330},[17192],{"type":21,"tag":322,"props":17193,"children":17194},{"emptyLinePlaceholder":3662},[17195],{"type":26,"value":3665},{"type":21,"tag":322,"props":17197,"children":17198},{"class":324,"line":5359},[17199,17203,17208],{"type":21,"tag":322,"props":17200,"children":17201},{"style":3638},[17202],{"type":26,"value":4798},{"type":21,"tag":322,"props":17204,"children":17205},{"style":4801},[17206],{"type":26,"value":17207}," qpe_phase_gate",{"type":21,"tag":322,"props":17209,"children":17210},{"style":329},[17211],{"type":26,"value":17212},"(n_count, theta):\n",{"type":21,"tag":322,"props":17214,"children":17215},{"class":324,"line":5368},[17216],{"type":21,"tag":322,"props":17217,"children":17218},{"style":359},[17219],{"type":26,"value":17220},"    \"\"\"QPE for U = P(2*pi*theta), whose eigenstate is |1>.\"\"\"\n",{"type":21,"tag":322,"props":17222,"children":17223},{"class":324,"line":5396},[17224,17228,17232,17237,17241,17245],{"type":21,"tag":322,"props":17225,"children":17226},{"style":329},[17227],{"type":26,"value":7934},{"type":21,"tag":322,"props":17229,"children":17230},{"style":3638},[17231],{"type":26,"value":3678},{"type":21,"tag":322,"props":17233,"children":17234},{"style":329},[17235],{"type":26,"value":17236}," QuantumCircuit(n_count ",{"type":21,"tag":322,"props":17238,"children":17239},{"style":3638},[17240],{"type":26,"value":6965},{"type":21,"tag":322,"props":17242,"children":17243},{"style":348},[17244],{"type":26,"value":6536},{"type":21,"tag":322,"props":17246,"children":17247},{"style":329},[17248],{"type":26,"value":17249},", n_count)\n",{"type":21,"tag":322,"props":17251,"children":17252},{"class":324,"line":5427},[17253,17257,17261,17266],{"type":21,"tag":322,"props":17254,"children":17255},{"style":329},[17256],{"type":26,"value":8326},{"type":21,"tag":322,"props":17258,"children":17259},{"style":348},[17260],{"type":26,"value":5379},{"type":21,"tag":322,"props":17262,"children":17263},{"style":329},[17264],{"type":26,"value":17265},"(n_count))             ",{"type":21,"tag":322,"props":17267,"children":17268},{"style":3888},[17269],{"type":26,"value":17270},"# counting register in superposition\n",{"type":21,"tag":322,"props":17272,"children":17273},{"class":324,"line":5449},[17274,17279],{"type":21,"tag":322,"props":17275,"children":17276},{"style":329},[17277],{"type":26,"value":17278},"    qc.x(n_count)                    ",{"type":21,"tag":322,"props":17280,"children":17281},{"style":3888},[17282],{"type":26,"value":17283},"# prepare the eigenstate |1>\n",{"type":21,"tag":322,"props":17285,"children":17286},{"class":324,"line":5495},[17287],{"type":21,"tag":322,"props":17288,"children":17289},{"emptyLinePlaceholder":3662},[17290],{"type":26,"value":3665},{"type":21,"tag":322,"props":17292,"children":17293},{"class":324,"line":5508},[17294,17298,17302,17306,17310,17315],{"type":21,"tag":322,"props":17295,"children":17296},{"style":3638},[17297],{"type":26,"value":5177},{"type":21,"tag":322,"props":17299,"children":17300},{"style":329},[17301],{"type":26,"value":13931},{"type":21,"tag":322,"props":17303,"children":17304},{"style":3638},[17305],{"type":26,"value":3941},{"type":21,"tag":322,"props":17307,"children":17308},{"style":348},[17309],{"type":26,"value":3946},{"type":21,"tag":322,"props":17311,"children":17312},{"style":329},[17313],{"type":26,"value":17314},"(n_count):         ",{"type":21,"tag":322,"props":17316,"children":17317},{"style":3888},[17318],{"type":26,"value":17319},"# controlled-U^(2^k) = CP(2*pi*theta*2^k)\n",{"type":21,"tag":322,"props":17321,"children":17322},{"class":324,"line":5516},[17323,17328,17332,17336,17341,17345,17350,17354,17358,17362],{"type":21,"tag":322,"props":17324,"children":17325},{"style":329},[17326],{"type":26,"value":17327},"        qc.cp(",{"type":21,"tag":322,"props":17329,"children":17330},{"style":348},[17331],{"type":26,"value":423},{"type":21,"tag":322,"props":17333,"children":17334},{"style":3638},[17335],{"type":26,"value":8782},{"type":21,"tag":322,"props":17337,"children":17338},{"style":329},[17339],{"type":26,"value":17340}," np.pi ",{"type":21,"tag":322,"props":17342,"children":17343},{"style":3638},[17344],{"type":26,"value":6830},{"type":21,"tag":322,"props":17346,"children":17347},{"style":329},[17348],{"type":26,"value":17349}," theta ",{"type":21,"tag":322,"props":17351,"children":17352},{"style":3638},[17353],{"type":26,"value":6830},{"type":21,"tag":322,"props":17355,"children":17356},{"style":348},[17357],{"type":26,"value":6821},{"type":21,"tag":322,"props":17359,"children":17360},{"style":3638},[17361],{"type":26,"value":8746},{"type":21,"tag":322,"props":17363,"children":17364},{"style":329},[17365],{"type":26,"value":17366}," k, k, n_count)\n",{"type":21,"tag":322,"props":17368,"children":17369},{"class":324,"line":5534},[17370],{"type":21,"tag":322,"props":17371,"children":17372},{"emptyLinePlaceholder":3662},[17373],{"type":26,"value":3665},{"type":21,"tag":322,"props":17375,"children":17376},{"class":324,"line":5552},[17377,17382,17386,17390,17394,17399,17403,17407,17411],{"type":21,"tag":322,"props":17378,"children":17379},{"style":329},[17380],{"type":26,"value":17381},"    qc.compose(qft(n_count).inverse(), ",{"type":21,"tag":322,"props":17383,"children":17384},{"style":3703},[17385],{"type":26,"value":1797},{"type":21,"tag":322,"props":17387,"children":17388},{"style":3638},[17389],{"type":26,"value":3678},{"type":21,"tag":322,"props":17391,"children":17392},{"style":348},[17393],{"type":26,"value":5379},{"type":21,"tag":322,"props":17395,"children":17396},{"style":329},[17397],{"type":26,"value":17398},"(n_count), ",{"type":21,"tag":322,"props":17400,"children":17401},{"style":3703},[17402],{"type":26,"value":8636},{"type":21,"tag":322,"props":17404,"children":17405},{"style":3638},[17406],{"type":26,"value":3678},{"type":21,"tag":322,"props":17408,"children":17409},{"style":348},[17410],{"type":26,"value":3715},{"type":21,"tag":322,"props":17412,"children":17413},{"style":329},[17414],{"type":26,"value":3738},{"type":21,"tag":322,"props":17416,"children":17417},{"class":324,"line":5560},[17418,17422,17426,17430,17434],{"type":21,"tag":322,"props":17419,"children":17420},{"style":329},[17421],{"type":26,"value":8691},{"type":21,"tag":322,"props":17423,"children":17424},{"style":348},[17425],{"type":26,"value":5379},{"type":21,"tag":322,"props":17427,"children":17428},{"style":329},[17429],{"type":26,"value":17398},{"type":21,"tag":322,"props":17431,"children":17432},{"style":348},[17433],{"type":26,"value":5379},{"type":21,"tag":322,"props":17435,"children":17436},{"style":329},[17437],{"type":26,"value":17438},"(n_count))\n",{"type":21,"tag":322,"props":17440,"children":17441},{"class":324,"line":5594},[17442,17446],{"type":21,"tag":322,"props":17443,"children":17444},{"style":3638},[17445],{"type":26,"value":5023},{"type":21,"tag":322,"props":17447,"children":17448},{"style":329},[17449],{"type":26,"value":8215},{"type":21,"tag":322,"props":17451,"children":17452},{"class":324,"line":5612},[17453],{"type":21,"tag":322,"props":17454,"children":17455},{"emptyLinePlaceholder":3662},[17456],{"type":26,"value":3665},{"type":21,"tag":322,"props":17458,"children":17459},{"class":324,"line":5637},[17460],{"type":21,"tag":322,"props":17461,"children":17462},{"emptyLinePlaceholder":3662},[17463],{"type":26,"value":3665},{"type":21,"tag":322,"props":17465,"children":17466},{"class":324,"line":5679},[17467,17471,17475],{"type":21,"tag":322,"props":17468,"children":17469},{"style":329},[17470],{"type":26,"value":8861},{"type":21,"tag":322,"props":17472,"children":17473},{"style":3638},[17474],{"type":26,"value":3678},{"type":21,"tag":322,"props":17476,"children":17477},{"style":329},[17478],{"type":26,"value":8870},{"type":21,"tag":322,"props":17480,"children":17481},{"class":324,"line":5687},[17482,17486,17490,17495,17499,17503,17508,17513],{"type":21,"tag":322,"props":17483,"children":17484},{"style":329},[17485],{"type":26,"value":7819},{"type":21,"tag":322,"props":17487,"children":17488},{"style":3638},[17489],{"type":26,"value":3678},{"type":21,"tag":322,"props":17491,"children":17492},{"style":329},[17493],{"type":26,"value":17494}," qpe_phase_gate(",{"type":21,"tag":322,"props":17496,"children":17497},{"style":348},[17498],{"type":26,"value":5969},{"type":21,"tag":322,"props":17500,"children":17501},{"style":329},[17502],{"type":26,"value":367},{"type":21,"tag":322,"props":17504,"children":17505},{"style":348},[17506],{"type":26,"value":17507},"0.625",{"type":21,"tag":322,"props":17509,"children":17510},{"style":329},[17511],{"type":26,"value":17512},")        ",{"type":21,"tag":322,"props":17514,"children":17515},{"style":3888},[17516],{"type":26,"value":17517},"# theta = 0.625 = 0.1010 in binary\n",{"type":21,"tag":322,"props":17519,"children":17520},{"class":324,"line":5722},[17521,17525,17529,17533,17537,17541,17545],{"type":21,"tag":322,"props":17522,"children":17523},{"style":329},[17524],{"type":26,"value":8894},{"type":21,"tag":322,"props":17526,"children":17527},{"style":3638},[17528],{"type":26,"value":3678},{"type":21,"tag":322,"props":17530,"children":17531},{"style":329},[17532],{"type":26,"value":8903},{"type":21,"tag":322,"props":17534,"children":17535},{"style":3703},[17536],{"type":26,"value":4560},{"type":21,"tag":322,"props":17538,"children":17539},{"style":3638},[17540],{"type":26,"value":3678},{"type":21,"tag":322,"props":17542,"children":17543},{"style":348},[17544],{"type":26,"value":8916},{"type":21,"tag":322,"props":17546,"children":17547},{"style":329},[17548],{"type":26,"value":5719},{"type":21,"tag":322,"props":17550,"children":17551},{"class":324,"line":5838},[17552],{"type":21,"tag":322,"props":17553,"children":17554},{"emptyLinePlaceholder":3662},[17555],{"type":26,"value":3665},{"type":21,"tag":322,"props":17557,"children":17558},{"class":324,"line":5846},[17559,17563,17568,17572,17577,17581,17585,17589,17593,17597,17601,17605],{"type":21,"tag":322,"props":17560,"children":17561},{"style":3638},[17562],{"type":26,"value":3931},{"type":21,"tag":322,"props":17564,"children":17565},{"style":329},[17566],{"type":26,"value":17567}," bits, c ",{"type":21,"tag":322,"props":17569,"children":17570},{"style":3638},[17571],{"type":26,"value":3941},{"type":21,"tag":322,"props":17573,"children":17574},{"style":348},[17575],{"type":26,"value":17576}," sorted",{"type":21,"tag":322,"props":17578,"children":17579},{"style":329},[17580],{"type":26,"value":13600},{"type":21,"tag":322,"props":17582,"children":17583},{"style":3703},[17584],{"type":26,"value":3979},{"type":21,"tag":322,"props":17586,"children":17587},{"style":3638},[17588],{"type":26,"value":3984},{"type":21,"tag":322,"props":17590,"children":17591},{"style":329},[17592],{"type":26,"value":13613},{"type":21,"tag":322,"props":17594,"children":17595},{"style":3638},[17596],{"type":26,"value":5285},{"type":21,"tag":322,"props":17598,"children":17599},{"style":329},[17600],{"type":26,"value":13622},{"type":21,"tag":322,"props":17602,"children":17603},{"style":348},[17604],{"type":26,"value":3994},{"type":21,"tag":322,"props":17606,"children":17607},{"style":329},[17608],{"type":26,"value":17609},"]):\n",{"type":21,"tag":322,"props":17611,"children":17612},{"class":324,"line":5891},[17613,17618,17622,17627,17632,17636,17640,17644,17648,17652],{"type":21,"tag":322,"props":17614,"children":17615},{"style":329},[17616],{"type":26,"value":17617},"    est ",{"type":21,"tag":322,"props":17619,"children":17620},{"style":3638},[17621],{"type":26,"value":3678},{"type":21,"tag":322,"props":17623,"children":17624},{"style":348},[17625],{"type":26,"value":17626}," int",{"type":21,"tag":322,"props":17628,"children":17629},{"style":329},[17630],{"type":26,"value":17631},"(bits, ",{"type":21,"tag":322,"props":17633,"children":17634},{"style":348},[17635],{"type":26,"value":423},{"type":21,"tag":322,"props":17637,"children":17638},{"style":329},[17639],{"type":26,"value":5784},{"type":21,"tag":322,"props":17641,"children":17642},{"style":3638},[17643],{"type":26,"value":174},{"type":21,"tag":322,"props":17645,"children":17646},{"style":348},[17647],{"type":26,"value":6821},{"type":21,"tag":322,"props":17649,"children":17650},{"style":3638},[17651],{"type":26,"value":8746},{"type":21,"tag":322,"props":17653,"children":17654},{"style":348},[17655],{"type":26,"value":12905},{"type":21,"tag":322,"props":17657,"children":17658},{"class":324,"line":5909},[17659,17664,17668,17672,17676,17680,17685,17689,17694,17699,17703,17707,17712,17716,17721,17725,17730,17734,17738,17743,17747,17752,17756,17760],{"type":21,"tag":322,"props":17660,"children":17661},{"style":348},[17662],{"type":26,"value":17663},"    print",{"type":21,"tag":322,"props":17665,"children":17666},{"style":329},[17667],{"type":26,"value":3758},{"type":21,"tag":322,"props":17669,"children":17670},{"style":3638},[17671],{"type":26,"value":3763},{"type":21,"tag":322,"props":17673,"children":17674},{"style":359},[17675],{"type":26,"value":3788},{"type":21,"tag":322,"props":17677,"children":17678},{"style":348},[17679],{"type":26,"value":3773},{"type":21,"tag":322,"props":17681,"children":17682},{"style":329},[17683],{"type":26,"value":17684},"bits",{"type":21,"tag":322,"props":17686,"children":17687},{"style":348},[17688],{"type":26,"value":3783},{"type":21,"tag":322,"props":17690,"children":17691},{"style":359},[17692],{"type":26,"value":17693}," -> theta = ",{"type":21,"tag":322,"props":17695,"children":17696},{"style":348},[17697],{"type":26,"value":17698},"{int",{"type":21,"tag":322,"props":17700,"children":17701},{"style":329},[17702],{"type":26,"value":17631},{"type":21,"tag":322,"props":17704,"children":17705},{"style":348},[17706],{"type":26,"value":423},{"type":21,"tag":322,"props":17708,"children":17709},{"style":329},[17710],{"type":26,"value":17711},")",{"type":21,"tag":322,"props":17713,"children":17714},{"style":348},[17715],{"type":26,"value":3783},{"type":21,"tag":322,"props":17717,"children":17718},{"style":359},[17719],{"type":26,"value":17720},"\u002F16 = ",{"type":21,"tag":322,"props":17722,"children":17723},{"style":348},[17724],{"type":26,"value":3773},{"type":21,"tag":322,"props":17726,"children":17727},{"style":329},[17728],{"type":26,"value":17729},"est",{"type":21,"tag":322,"props":17731,"children":17732},{"style":3638},[17733],{"type":26,"value":4085},{"type":21,"tag":322,"props":17735,"children":17736},{"style":348},[17737],{"type":26,"value":3783},{"type":21,"tag":322,"props":17739,"children":17740},{"style":359},[17741],{"type":26,"value":17742},"   counts = ",{"type":21,"tag":322,"props":17744,"children":17745},{"style":348},[17746],{"type":26,"value":3773},{"type":21,"tag":322,"props":17748,"children":17749},{"style":329},[17750],{"type":26,"value":17751},"c",{"type":21,"tag":322,"props":17753,"children":17754},{"style":348},[17755],{"type":26,"value":3783},{"type":21,"tag":322,"props":17757,"children":17758},{"style":359},[17759],{"type":26,"value":3788},{"type":21,"tag":322,"props":17761,"children":17762},{"style":329},[17763],{"type":26,"value":3738},{"type":21,"tag":22,"props":17765,"children":17766},{},[17767],{"type":26,"value":12823},{"type":21,"tag":312,"props":17769,"children":17771},{"code":17770},"1010 -> theta = 10\u002F16 = 0.6250   counts = 4096\n",[17772],{"type":21,"tag":79,"props":17773,"children":17774},{"__ignoreMap":7},[17775],{"type":26,"value":17770},{"type":21,"tag":22,"props":17777,"children":17778},{},[17779,17784],{"type":21,"tag":71,"props":17780,"children":17781},{},[17782],{"type":26,"value":17783},"4096 out of 4096 shots.",{"type":26,"value":17785}," θ = 0.625 is exactly representable in 4 bits (0.1010₂ = 1\u002F2 + 1\u002F8), so QPE returns it with certainty. Three counting qubits and θ = 0.125 behaves the same way:",{"type":21,"tag":312,"props":17787,"children":17789},{"code":17788},"[('001', 4096)]\n",[17790],{"type":21,"tag":79,"props":17791,"children":17792},{"__ignoreMap":7},[17793],{"type":26,"value":17788},{"type":21,"tag":22,"props":17795,"children":17796},{},[17797],{"type":26,"value":17798},"001 = 1, and 1\u002F8 = 0.125. Exact again.",{"type":21,"tag":17800,"props":17801,"children":17803},"h3",{"id":17802},"the-endianness-trap",[17804],{"type":26,"value":17805},"The endianness trap",{"type":21,"tag":22,"props":17807,"children":17808},{},[17809,17815,17817,17822,17824,17828,17830,17836],{"type":21,"tag":79,"props":17810,"children":17812},{"className":17811},[],[17813],{"type":26,"value":17814},"int(bits, 2) \u002F 2**n",{"type":26,"value":17816}," worked above, and it is worth understanding ",{"type":21,"tag":51,"props":17818,"children":17819},{},[17820],{"type":26,"value":17821},"why",{"type":26,"value":17823}," rather than copying it. Qiskit prints bitstrings with classical bit 0 on the ",{"type":21,"tag":71,"props":17825,"children":17826},{},[17827],{"type":26,"value":8225},{"type":26,"value":17829},", and our counting qubit 0 (the one that got the smallest rotation, ",{"type":21,"tag":79,"props":17831,"children":17833},{"className":17832},[],[17834],{"type":26,"value":17835},"U^1",{"type":26,"value":17837},") is the least significant bit of j. The QFT's final swap network is what lines those two conventions up.",{"type":21,"tag":22,"props":17839,"children":17840},{},[17841,17843,17849],{"type":26,"value":17842},"Drop the swaps from ",{"type":21,"tag":79,"props":17844,"children":17846},{"className":17845},[],[17847],{"type":26,"value":17848},"qft()",{"type":26,"value":17850}," (a tempting \"optimization\", since they look cosmetic) and rerun the exact θ = 0.625 case. Real output over 2048 shots:",{"type":21,"tag":312,"props":17852,"children":17854},{"code":17853},"{'1101': 1007, '0101': 470, '0001': 251, '0100': 80, '1100': 77,\n '0010': 66, '1110': 64, '0110': 16, '1010': 9, '1001': 8}\n",[17855],{"type":21,"tag":79,"props":17856,"children":17857},{"__ignoreMap":7},[17858],{"type":26,"value":17853},{"type":21,"tag":22,"props":17860,"children":17861},{},[17862,17864,17870,17872,17877],{"type":26,"value":17863},"The correct answer ",{"type":21,"tag":79,"props":17865,"children":17867},{"className":17866},[],[17868],{"type":26,"value":17869},"1010",{"type":26,"value":17871}," now gets ",{"type":21,"tag":71,"props":17873,"children":17874},{},[17875],{"type":26,"value":17876},"9 shots out of 2048",{"type":26,"value":17878},". The distribution smeared across the register with its largest peak at 1101, a confidently wrong 0.8125. Nothing errors. Nothing warns. You publish the wrong number, and with a θ you didn't already know you would have no way to tell. This is the same silent-failure class as the endianness bug in the Grover oracle, and it's why the QFT post insisted on keeping the swaps in a reusable block.",{"type":21,"tag":22,"props":17880,"children":17881},{},[17882],{"type":26,"value":17883},"Always sanity-check a QPE implementation against a θ you already know before pointing it at one you don't.",{"type":21,"tag":59,"props":17885,"children":17887},{"id":17886},"when-θ-doesnt-fit-in-n-bits",[17888],{"type":26,"value":17889},"When θ doesn't fit in n bits",{"type":21,"tag":22,"props":17891,"children":17892},{},[17893],{"type":26,"value":17894},"Real eigenphases are not tidy binary fractions. Try θ = 0.3, which has an infinite binary expansion, with 4 counting qubits:",{"type":21,"tag":312,"props":17896,"children":17898},{"code":17897,"language":3628,"meta":7,"className":3626,"style":7},"qc = qpe_phase_gate(4, 0.3)\ncounts = sim.run(transpile(qc, sim), shots=4096).result().get_counts()\nfor bits, c in sorted(counts.items(), key=lambda kv: -kv[1])[:6]:\n    print(f\"{bits}  theta_est = {int(bits, 2) \u002F 2**4:.4f}   p = {c \u002F 4096:.4f}\")\n",[17899],{"type":21,"tag":79,"props":17900,"children":17901},{"__ignoreMap":7},[17902,17933,17964,18025],{"type":21,"tag":322,"props":17903,"children":17904},{"class":324,"line":325},[17905,17909,17913,17917,17921,17925,17929],{"type":21,"tag":322,"props":17906,"children":17907},{"style":329},[17908],{"type":26,"value":7819},{"type":21,"tag":322,"props":17910,"children":17911},{"style":3638},[17912],{"type":26,"value":3678},{"type":21,"tag":322,"props":17914,"children":17915},{"style":329},[17916],{"type":26,"value":17494},{"type":21,"tag":322,"props":17918,"children":17919},{"style":348},[17920],{"type":26,"value":5969},{"type":21,"tag":322,"props":17922,"children":17923},{"style":329},[17924],{"type":26,"value":367},{"type":21,"tag":322,"props":17926,"children":17927},{"style":348},[17928],{"type":26,"value":15180},{"type":21,"tag":322,"props":17930,"children":17931},{"style":329},[17932],{"type":26,"value":3738},{"type":21,"tag":322,"props":17934,"children":17935},{"class":324,"line":335},[17936,17940,17944,17948,17952,17956,17960],{"type":21,"tag":322,"props":17937,"children":17938},{"style":329},[17939],{"type":26,"value":8894},{"type":21,"tag":322,"props":17941,"children":17942},{"style":3638},[17943],{"type":26,"value":3678},{"type":21,"tag":322,"props":17945,"children":17946},{"style":329},[17947],{"type":26,"value":8903},{"type":21,"tag":322,"props":17949,"children":17950},{"style":3703},[17951],{"type":26,"value":4560},{"type":21,"tag":322,"props":17953,"children":17954},{"style":3638},[17955],{"type":26,"value":3678},{"type":21,"tag":322,"props":17957,"children":17958},{"style":348},[17959],{"type":26,"value":8916},{"type":21,"tag":322,"props":17961,"children":17962},{"style":329},[17963],{"type":26,"value":5719},{"type":21,"tag":322,"props":17965,"children":17966},{"class":324,"line":344},[17967,17971,17975,17979,17983,17987,17991,17995,17999,18003,18007,18011,18016,18020],{"type":21,"tag":322,"props":17968,"children":17969},{"style":3638},[17970],{"type":26,"value":3931},{"type":21,"tag":322,"props":17972,"children":17973},{"style":329},[17974],{"type":26,"value":17567},{"type":21,"tag":322,"props":17976,"children":17977},{"style":3638},[17978],{"type":26,"value":3941},{"type":21,"tag":322,"props":17980,"children":17981},{"style":348},[17982],{"type":26,"value":17576},{"type":21,"tag":322,"props":17984,"children":17985},{"style":329},[17986],{"type":26,"value":13600},{"type":21,"tag":322,"props":17988,"children":17989},{"style":3703},[17990],{"type":26,"value":3979},{"type":21,"tag":322,"props":17992,"children":17993},{"style":3638},[17994],{"type":26,"value":3984},{"type":21,"tag":322,"props":17996,"children":17997},{"style":329},[17998],{"type":26,"value":13613},{"type":21,"tag":322,"props":18000,"children":18001},{"style":3638},[18002],{"type":26,"value":5285},{"type":21,"tag":322,"props":18004,"children":18005},{"style":329},[18006],{"type":26,"value":13622},{"type":21,"tag":322,"props":18008,"children":18009},{"style":348},[18010],{"type":26,"value":3994},{"type":21,"tag":322,"props":18012,"children":18013},{"style":329},[18014],{"type":26,"value":18015},"])[:",{"type":21,"tag":322,"props":18017,"children":18018},{"style":348},[18019],{"type":26,"value":14116},{"type":21,"tag":322,"props":18021,"children":18022},{"style":329},[18023],{"type":26,"value":18024},"]:\n",{"type":21,"tag":322,"props":18026,"children":18027},{"class":324,"line":407},[18028,18032,18036,18040,18044,18048,18052,18056,18061,18065,18069,18073,18077,18081,18085,18089,18093,18097,18101,18105,18109,18114,18118,18123,18127,18131,18135],{"type":21,"tag":322,"props":18029,"children":18030},{"style":348},[18031],{"type":26,"value":17663},{"type":21,"tag":322,"props":18033,"children":18034},{"style":329},[18035],{"type":26,"value":3758},{"type":21,"tag":322,"props":18037,"children":18038},{"style":3638},[18039],{"type":26,"value":3763},{"type":21,"tag":322,"props":18041,"children":18042},{"style":359},[18043],{"type":26,"value":3788},{"type":21,"tag":322,"props":18045,"children":18046},{"style":348},[18047],{"type":26,"value":3773},{"type":21,"tag":322,"props":18049,"children":18050},{"style":329},[18051],{"type":26,"value":17684},{"type":21,"tag":322,"props":18053,"children":18054},{"style":348},[18055],{"type":26,"value":3783},{"type":21,"tag":322,"props":18057,"children":18058},{"style":359},[18059],{"type":26,"value":18060},"  theta_est = ",{"type":21,"tag":322,"props":18062,"children":18063},{"style":348},[18064],{"type":26,"value":17698},{"type":21,"tag":322,"props":18066,"children":18067},{"style":329},[18068],{"type":26,"value":17631},{"type":21,"tag":322,"props":18070,"children":18071},{"style":348},[18072],{"type":26,"value":423},{"type":21,"tag":322,"props":18074,"children":18075},{"style":329},[18076],{"type":26,"value":5784},{"type":21,"tag":322,"props":18078,"children":18079},{"style":3638},[18080],{"type":26,"value":174},{"type":21,"tag":322,"props":18082,"children":18083},{"style":348},[18084],{"type":26,"value":6821},{"type":21,"tag":322,"props":18086,"children":18087},{"style":3638},[18088],{"type":26,"value":5150},{"type":21,"tag":322,"props":18090,"children":18091},{"style":348},[18092],{"type":26,"value":5969},{"type":21,"tag":322,"props":18094,"children":18095},{"style":3638},[18096],{"type":26,"value":4085},{"type":21,"tag":322,"props":18098,"children":18099},{"style":348},[18100],{"type":26,"value":3783},{"type":21,"tag":322,"props":18102,"children":18103},{"style":359},[18104],{"type":26,"value":13404},{"type":21,"tag":322,"props":18106,"children":18107},{"style":348},[18108],{"type":26,"value":3773},{"type":21,"tag":322,"props":18110,"children":18111},{"style":329},[18112],{"type":26,"value":18113},"c ",{"type":21,"tag":322,"props":18115,"children":18116},{"style":3638},[18117],{"type":26,"value":174},{"type":21,"tag":322,"props":18119,"children":18120},{"style":348},[18121],{"type":26,"value":18122}," 4096",{"type":21,"tag":322,"props":18124,"children":18125},{"style":3638},[18126],{"type":26,"value":4085},{"type":21,"tag":322,"props":18128,"children":18129},{"style":348},[18130],{"type":26,"value":3783},{"type":21,"tag":322,"props":18132,"children":18133},{"style":359},[18134],{"type":26,"value":3788},{"type":21,"tag":322,"props":18136,"children":18137},{"style":329},[18138],{"type":26,"value":3738},{"type":21,"tag":22,"props":18140,"children":18141},{},[18142],{"type":26,"value":12823},{"type":21,"tag":312,"props":18144,"children":18146},{"code":18145},"0101  theta_est = 0.3125   p = 0.8823\n0100  theta_est = 0.2500   p = 0.0542\n0110  theta_est = 0.3750   p = 0.0220\n0011  theta_est = 0.1875   p = 0.0093\n0111  theta_est = 0.4375   p = 0.0063\n0010  theta_est = 0.1250   p = 0.0049\n",[18147],{"type":21,"tag":79,"props":18148,"children":18149},{"__ignoreMap":7},[18150],{"type":26,"value":18145},{"type":21,"tag":22,"props":18152,"children":18153},{},[18154,18156,18161],{"type":26,"value":18155},"The distribution is peaked, not delta-shaped. 88% of shots land on 5\u002F16 = 0.3125, the closest 4-bit value to 0.3. The rest spread out, decaying fast as you move away from the peak. This is the generic behaviour: ",{"type":21,"tag":71,"props":18157,"children":18158},{},[18159],{"type":26,"value":18160},"you get the nearest n-bit value with high probability, not certainty.",{"type":26,"value":18162}," The standard guarantee is that the best n-bit estimate appears with probability at least 4\u002Fπ² ≈ 40.5%. Here we're comfortably above that.",{"type":21,"tag":22,"props":18164,"children":18165},{},[18166,18168,18173],{"type":26,"value":18167},"Two things improve together when you add counting qubits: finer resolution ",{"type":21,"tag":51,"props":18169,"children":18170},{},[18171],{"type":26,"value":18172},"and",{"type":26,"value":18174}," a tighter peak. Same θ = 0.3 with 8 counting qubits:",{"type":21,"tag":312,"props":18176,"children":18178},{"code":18177},"01001101  theta_est = 0.30078   p = 0.8718\n01001100  theta_est = 0.29688   p = 0.0596\n01001110  theta_est = 0.30469   p = 0.0276\n01001111  theta_est = 0.30859   p = 0.0076\n01001011  theta_est = 0.29297   p = 0.0073\n",[18179],{"type":21,"tag":79,"props":18180,"children":18181},{"__ignoreMap":7},[18182],{"type":26,"value":18177},{"type":21,"tag":22,"props":18184,"children":18185},{},[18186,18188,18193],{"type":26,"value":18187},"Resolution went from ±0.031 to ±0.002, and the peak stayed at 87%. In general, to get n bits of precision with success probability 1 − ε you use n + O(log(1\u002Fε)) counting qubits. A handful of extra ",{"type":21,"tag":29,"props":18189,"children":18190},{"href":9414},[18191],{"type":26,"value":18192},"ancillas",{"type":26,"value":18194}," buys you a lot of confidence.",{"type":21,"tag":22,"props":18196,"children":18197},{},[18198,18200,18205],{"type":26,"value":18199},"The cost is the part people underestimate. Counting qubit k needs U applied 2^k times. Total controlled-U applications across the register: 2ⁿ − 1. Every bit of precision ",{"type":21,"tag":71,"props":18201,"children":18202},{},[18203],{"type":26,"value":18204},"doubles",{"type":26,"value":18206}," the circuit depth. That's fine when U^(2^k) has an efficient closed form. For a phase gate it's only one gate with a bigger angle, which is why the demo above is cheap. For modular exponentiation in Shor's it's a genuine arithmetic circuit repeated exponentially many times, and the depth is brutal.",{"type":21,"tag":59,"props":18208,"children":18210},{"id":18209},"where-qpe-gets-used",[18211],{"type":26,"value":18212},"Where QPE gets used",{"type":21,"tag":22,"props":18214,"children":18215},{},[18216,18221,18223,18227,18229,18234],{"type":21,"tag":71,"props":18217,"children":18218},{},[18219],{"type":26,"value":18220},"Shor's algorithm.",{"type":26,"value":18222}," Take U|y⟩ = |ay mod N⟩. Its eigenphases are s\u002Fr for integer s, where r is the multiplicative order of a mod N. Run QPE, measure s\u002Fr as a binary fraction, recover r by continued fractions, and factor N with a GCD. That's the whole quantum part. Everything else in ",{"type":21,"tag":29,"props":18224,"children":18225},{"href":9045},[18226],{"type":26,"value":10086},{"type":26,"value":18228}," is number theory. Order finding is QPE. QPE is controlled powers plus an inverse QFT. The chain from the QFT post now closes. (The algorithm was recently ",{"type":21,"tag":29,"props":18230,"children":18231},{"href":10097},[18232],{"type":26,"value":18233},"machine-verified in Lean",{"type":26,"value":18235},", if you want the formal treatment.)",{"type":21,"tag":22,"props":18237,"children":18238},{},[18239,18244,18246,18251],{"type":21,"tag":71,"props":18240,"children":18241},{},[18242],{"type":26,"value":18243},"Quantum chemistry.",{"type":26,"value":18245}," Set U = e^{-iHt}. QPE returns the ground-state energy to whatever precision you're willing to pay depth for, and crucially, it returns it with a ",{"type":21,"tag":51,"props":18247,"children":18248},{},[18249],{"type":26,"value":18250},"provable",{"type":26,"value":18252}," error bound rather than a variational estimate. This is the long-term goal of quantum simulation, and it's the application most likely to matter commercially before cryptanalysis does.",{"type":21,"tag":22,"props":18254,"children":18255},{},[18256,18261,18262,18266,18268,18272,18274,18279],{"type":21,"tag":71,"props":18257,"children":18258},{},[18259],{"type":26,"value":18260},"Contrast with VQE.",{"type":26,"value":1901},{"type":21,"tag":29,"props":18263,"children":18264},{"href":922},[18265],{"type":26,"value":883},{"type":26,"value":18267}," attacks the same chemistry problem from the opposite direction. Instead of one deep coherent circuit, it uses a shallow ",{"type":21,"tag":29,"props":18269,"children":18270},{"href":14271},[18271],{"type":26,"value":14274},{"type":26,"value":18273}," plus a classical optimizer, as introduced in ",{"type":21,"tag":29,"props":18275,"children":18276},{"href":929},[18277],{"type":26,"value":18278},"Peruzzo et al. 2014",{"type":26,"value":18280},". The trade is explicit:",{"type":21,"tag":4503,"props":18282,"children":18283},{},[18284,18302],{"type":21,"tag":4507,"props":18285,"children":18286},{},[18287],{"type":21,"tag":4511,"props":18288,"children":18289},{},[18290,18293,18298],{"type":21,"tag":4515,"props":18291,"children":18292},{},[],{"type":21,"tag":4515,"props":18294,"children":18295},{},[18296],{"type":26,"value":18297},"QPE",{"type":21,"tag":4515,"props":18299,"children":18300},{},[18301],{"type":26,"value":883},{"type":21,"tag":4529,"props":18303,"children":18304},{},[18305,18323,18341,18359],{"type":21,"tag":4511,"props":18306,"children":18307},{},[18308,18313,18318],{"type":21,"tag":4536,"props":18309,"children":18310},{},[18311],{"type":26,"value":18312},"Circuit depth",{"type":21,"tag":4536,"props":18314,"children":18315},{},[18316],{"type":26,"value":18317},"Deep, grows as 2ⁿ in precision bits",{"type":21,"tag":4536,"props":18319,"children":18320},{},[18321],{"type":26,"value":18322},"Shallow, fixed ansatz",{"type":21,"tag":4511,"props":18324,"children":18325},{},[18326,18331,18336],{"type":21,"tag":4536,"props":18327,"children":18328},{},[18329],{"type":26,"value":18330},"Precision",{"type":21,"tag":4536,"props":18332,"children":18333},{},[18334],{"type":26,"value":18335},"Provable, arbitrary",{"type":21,"tag":4536,"props":18337,"children":18338},{},[18339],{"type":26,"value":18340},"Heuristic, ansatz-limited",{"type":21,"tag":4511,"props":18342,"children":18343},{},[18344,18349,18354],{"type":21,"tag":4536,"props":18345,"children":18346},{},[18347],{"type":26,"value":18348},"Measurements",{"type":21,"tag":4536,"props":18350,"children":18351},{},[18352],{"type":26,"value":18353},"Few shots",{"type":21,"tag":4536,"props":18355,"children":18356},{},[18357],{"type":26,"value":18358},"Many thousands per iteration",{"type":21,"tag":4511,"props":18360,"children":18361},{},[18362,18367,18372],{"type":21,"tag":4536,"props":18363,"children":18364},{},[18365],{"type":26,"value":18366},"Hardware era",{"type":21,"tag":4536,"props":18368,"children":18369},{},[18370],{"type":26,"value":18371},"Fault-tolerant",{"type":21,"tag":4536,"props":18373,"children":18374},{},[18375],{"type":21,"tag":29,"props":18376,"children":18377},{"href":263},[18378],{"type":26,"value":266},{"type":21,"tag":22,"props":18380,"children":18381},{},[18382,18384,18389,18391,18395,18397,18401],{"type":26,"value":18383},"VQE runs today and gives you an answer you can't fully trust. QPE gives you an answer you trust and can't run today. Our ",{"type":21,"tag":29,"props":18385,"children":18386},{"href":893},[18387],{"type":26,"value":18388},"VQE walkthrough in PennyLane",{"type":26,"value":18390}," covers the near-term side. ",{"type":21,"tag":29,"props":18392,"children":18393},{"href":14507},[18394],{"type":26,"value":14223},{"type":26,"value":18396}," is generally the better fit for variational work while ",{"type":21,"tag":29,"props":18398,"children":18399},{"href":7528},[18400],{"type":26,"value":1962},{"type":26,"value":18402}," suits structured circuits like this one.",{"type":21,"tag":59,"props":18404,"children":18406},{"id":18405},"the-honest-caveat",[18407],{"type":26,"value":18408},"The honest caveat",{"type":21,"tag":22,"props":18410,"children":18411},{},[18412,18414,18418],{"type":26,"value":18413},"QPE is not a NISQ algorithm and no amount of clever ",{"type":21,"tag":29,"props":18415,"children":18416},{"href":6192},[18417],{"type":26,"value":6195},{"type":26,"value":18419}," will make it one.",{"type":21,"tag":22,"props":18421,"children":18422},{},[18423,18425,18429],{"type":26,"value":18424},"The circuit must stay coherent through 2ⁿ − 1 controlled-U applications, each of which decomposes into many native two-qubit gates. With two-qubit error around 0.5% on current hardware, a few hundred entangling gates leaves you with noise. ",{"type":21,"tag":29,"props":18426,"children":18427},{"href":9366},[18428],{"type":26,"value":16256},{"type":26,"value":18430}," sets a hard wall long before you reach chemically useful precision. Published hardware demonstrations of QPE are toy instances with 2-4 counting qubits, real, but not computations anyone needed done.",{"type":21,"tag":22,"props":18432,"children":18433},{},[18434,18436,18440,18442,18447,18449,18453,18455,18459,18461,18465,18467,18472],{"type":26,"value":18435},"QPE becomes practical when circuits run on ",{"type":21,"tag":29,"props":18437,"children":18438},{"href":1983},[18439],{"type":26,"value":1986},{"type":26,"value":18441}," rather than physical ones. Google's ",{"type":21,"tag":29,"props":18443,"children":18444},{"href":4184},[18445],{"type":26,"value":18446},"below-threshold result",{"type":26,"value":18448}," is the first real evidence that path is open, and our ",{"type":21,"tag":29,"props":18450,"children":18451},{"href":751},[18452],{"type":26,"value":2096},{"type":26,"value":18454}," covers what still has to happen. Until then: develop QPE on ",{"type":21,"tag":29,"props":18456,"children":18457},{"href":1005},[18458],{"type":26,"value":9211},{"type":26,"value":18460},", where you verify it exactly, and reach for VQE or ",{"type":21,"tag":29,"props":18462,"children":18463},{"href":4423},[18464],{"type":26,"value":9716},{"type":26,"value":18466}," if you need results from ",{"type":21,"tag":29,"props":18468,"children":18469},{"href":806},[18470],{"type":26,"value":18471},"today's hardware",{"type":26,"value":534},{"type":21,"tag":22,"props":18474,"children":18475},{},[18476],{"type":26,"value":18477},"That's not a reason to skip learning it. QPE is the template for what a fault-tolerant algorithm looks like (deep, exact, and structured) and it's the piece that makes the QFT worth building.",{"type":21,"tag":59,"props":18479,"children":18480},{"id":9223},[18481],{"type":26,"value":18482},"Next steps",{"type":21,"tag":3556,"props":18484,"children":18485},{},[18486,18498,18511,18535],{"type":21,"tag":3560,"props":18487,"children":18488},{},[18489,18496],{"type":21,"tag":71,"props":18490,"children":18491},{},[18492],{"type":21,"tag":29,"props":18493,"children":18494},{"href":814},[18495],{"type":26,"value":9265},{"type":26,"value":18497},": build the same QPE circuit in Cirq or Braket",{"type":21,"tag":3560,"props":18499,"children":18500},{},[18501,18509],{"type":21,"tag":71,"props":18502,"children":18503},{},[18504],{"type":21,"tag":29,"props":18505,"children":18506},{"href":4314},[18507],{"type":26,"value":18508},"Courses",{"type":26,"value":18510},": structured paths through the algorithm canon",{"type":21,"tag":3560,"props":18512,"children":18513},{},[18514,18521,18522,18528,18529,18534],{"type":21,"tag":71,"props":18515,"children":18516},{},[18517],{"type":21,"tag":29,"props":18518,"children":18519},{"href":4400},[18520],{"type":26,"value":9286},{"type":26,"value":418},{"type":21,"tag":29,"props":18523,"children":18525},{"href":18524},"\u002Fglossary\u002Fquantum-gate",[18526],{"type":26,"value":18527},"quantum gate",{"type":26,"value":367},{"type":21,"tag":29,"props":18530,"children":18531},{"href":9414},[18532],{"type":26,"value":18533},"ancilla qubit",{"type":26,"value":9302},{"type":21,"tag":3560,"props":18536,"children":18537},{},[18538],{"type":26,"value":18539},"Extend the code above to a 2-qubit U with two eigenstates, feed in a superposition of both, and watch the counting register entangle with the eigenstate register. That's the mechanism behind order finding.",{"type":21,"tag":703,"props":18541,"children":18542},{},[18543],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":18545},[18546,18547,18548,18549,18552,18553,18554,18555],{"id":16515,"depth":335,"text":16518},{"id":16606,"depth":335,"text":16609},{"id":16711,"depth":335,"text":16714},{"id":12203,"depth":335,"text":12206,"children":18550},[18551],{"id":17802,"depth":344,"text":17805},{"id":17886,"depth":335,"text":17889},{"id":18209,"depth":335,"text":18212},{"id":18405,"depth":335,"text":18408},{"id":9223,"depth":335,"text":18482},"content:blog:quantum-phase-estimation-tutorial.md","blog\u002Fquantum-phase-estimation-tutorial.md","blog\u002Fquantum-phase-estimation-tutorial",{"_path":18560,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":18561,"description":18562,"date":18563,"author":11,"tags":18564,"readingTime":16,"body":18566,"_type":715,"_id":19580,"_source":717,"_file":19581,"_stem":19582,"_extension":720},"\u002Fja\u002Fblog\u002Fquantum-computing-new-vocabulary","量子コンピューティングの新しい語彙: 主流になる16の用語","AIが私たちに『トークン』『ハルシネーション』『ファインチューニング』を与えたように、量子コンピューティングも独自の語彙を築き上げつつあります。ここでは、数年以内に開発者や企業が使うようになる用語を紹介します。","2026-03-15",[7586,1033,18565],"Terminology",{"type":18,"children":18567,"toc":19561},[18568,18605,18610,18615,18619,18625,18633,18638,18664,18669,18672,18678,18686,18705,18710,18718,18723,18726,18732,18740,18752,18763,18775,18778,18784,18792,18803,18815,18818,18824,18832,18844,18934,18939,18942,18948,18956,18981,18993,19001,19006,19009,19015,19023,19035,19040,19043,19049,19057,19066,19071,19074,19080,19088,19112,19124,19129,19134,19137,19143,19151,19163,19175,19187,19190,19196,19204,19216,19221,19224,19230,19238,19254,19259,19262,19268,19276,19285,19290,19293,19299,19307,19326,19331,19334,19340,19348,19353,19363,19379,19384,19387,19393,19401,19413,19436,19447,19450,19455,19460,19533,19538,19543,19557],{"type":21,"tag":22,"props":18569,"children":18570},{},[18571,18573,18578,18580,18585,18586,18591,18592,18597,18598,18603],{"type":26,"value":18572},"ディープラーニングが2022〜2023年頃に主流へと躍り出たとき、それはまったく新しい語彙を日常の会話へと引きずり込みました。突如として誰もが ",{"type":21,"tag":51,"props":18574,"children":18575},{},[18576],{"type":26,"value":18577},"tokens",{"type":26,"value":18579},"、",{"type":21,"tag":51,"props":18581,"children":18582},{},[18583],{"type":26,"value":18584},"hallucinations",{"type":26,"value":18579},{"type":21,"tag":51,"props":18587,"children":18588},{},[18589],{"type":26,"value":18590},"context windows",{"type":26,"value":18579},{"type":21,"tag":51,"props":18593,"children":18594},{},[18595],{"type":26,"value":18596},"fine-tuning",{"type":26,"value":18579},{"type":21,"tag":51,"props":18599,"children":18600},{},[18601],{"type":26,"value":18602},"prompt engineering",{"type":26,"value":18604}," について語り始めたのです。これらは10年前には一般的な開発者の語彙にほとんど存在しなかった用語です。新世代の「AIエンジニア」が登場し、GPU時間は貴重な資源となり、「推論コスト」はエンジニアリング予算の項目になりました。",{"type":21,"tag":22,"props":18606,"children":18607},{},[18608],{"type":26,"value":18609},"量子コンピューティングも同様の軌道をたどっています。曲線上ではまだ早い段階にありますが、変曲点は近づいています。そしてそれとともに、研究論文から求人票、スタートアップのピッチ、そしてStack Overflowの質問へと移っていく新しい用語群が現れます。これらの中には、すでに量子コミュニティで存在するものもあります。ほとんどはその外では知られていません。しかしそのすべてが、他の誰よりも先に、今のうちに学ぶ価値があります。",{"type":21,"tag":22,"props":18611,"children":18612},{},[18613],{"type":26,"value":18614},"ここでは、主流になる態勢が整っている量子コンピューティングの世界の16の用語を紹介します。",{"type":21,"tag":18616,"props":18617,"children":18618},"hr",{},[],{"type":21,"tag":59,"props":18620,"children":18622},{"id":18621},"_1-qpu-quantum-processing-unit",[18623],{"type":26,"value":18624},"1. QPU (Quantum Processing Unit)",{"type":21,"tag":22,"props":18626,"children":18627},{},[18628],{"type":21,"tag":71,"props":18629,"children":18630},{},[18631],{"type":26,"value":18632},"AIとの対応: GPU",{"type":21,"tag":22,"props":18634,"children":18635},{},[18636],{"type":26,"value":18637},"「GPUへのアクセスはある?」は、MLチームで標準的な質問になりました。「QPUへのアクセスはある?」は、その量子版になるでしょう。",{"type":21,"tag":22,"props":18639,"children":18640},{},[18641,18643,18648,18650,18655,18657,18662],{"type":26,"value":18642},"QPUは量子回路を実行するハードウェアチップです。IBMは自社のシステムを ",{"type":21,"tag":51,"props":18644,"children":18645},{},[18646],{"type":26,"value":18647},"quantum computers",{"type":26,"value":18649}," と呼び、NVIDIAはGPUアクセラレーテッドなシミュレーションのための ",{"type":21,"tag":71,"props":18651,"children":18652},{},[18653],{"type":26,"value":18654},"CUDA-Q",{"type":26,"value":18656}," を持ち、IonQのトラップイオン方式のシステムは今日利用可能な中で最も高忠実度のQPUです。「GPU時間」がエンジニアたちが奪い合う資源になったのと同じように、",{"type":21,"tag":71,"props":18658,"children":18659},{},[18660],{"type":26,"value":18661},"QPU時間",{"type":26,"value":18663}," — 量子ハードウェア上での実際の実行サイクル — は、プレミアムでスケジュール可能な資源になるでしょう。",{"type":21,"tag":22,"props":18665,"children":18666},{},[18667],{"type":26,"value":18668},"この用語はすでにクラウドの価格ページ (AWS Braket、Azure Quantum) に登場しています。5年以内には求人票に載るでしょう。",{"type":21,"tag":18616,"props":18670,"children":18671},{},[],{"type":21,"tag":59,"props":18673,"children":18675},{"id":18674},"_2-shot-budget",[18676],{"type":26,"value":18677},"2. Shot Budget",{"type":21,"tag":22,"props":18679,"children":18680},{},[18681],{"type":21,"tag":71,"props":18682,"children":18683},{},[18684],{"type":26,"value":18685},"AIとの対応: トークン予算 \u002F コンテキストウィンドウ",{"type":21,"tag":22,"props":18687,"children":18688},{},[18689,18691,18696,18698,18703],{"type":26,"value":18690},"量子回路を実行するとき、あなたは単一の答えを得るのではありません。回路を何度も実行することでサンプリングされた ",{"type":21,"tag":51,"props":18692,"children":18693},{},[18694],{"type":26,"value":18695},"確率分布",{"type":26,"value":18697}," を得るのです。1回の実行は ",{"type":21,"tag":71,"props":18699,"children":18700},{},[18701],{"type":26,"value":18702},"shot",{"type":26,"value":18704}," と呼ばれます。1,000回実行される回路は、同じ回路を100回実行するよりも多くのQPU時間を要します。",{"type":21,"tag":22,"props":18706,"children":18707},{},[18708],{"type":26,"value":18709},"「shot budget」 — 自分が支払える実行回数 — は、QPUアクセスが拡大するにつれて現実的な最適化の関心事になるでしょう。開発者がトークン制限内に収めるためにプロンプトを削ることを学んだのと同じように、量子開発者は最小のshot budgetから最大の信号を引き出す回路を設計することを学ぶでしょう。",{"type":21,"tag":200,"props":18711,"children":18712},{},[18713],{"type":21,"tag":22,"props":18714,"children":18715},{},[18716],{"type":26,"value":18717},"「VQEを2,000ではなく200 shotで収束させました。これでQPUコストが90%削減されました。」",{"type":21,"tag":22,"props":18719,"children":18720},{},[18721],{"type":26,"value":18722},"これはすでにVQEやQAOAのような変分アルゴリズムにおける現実的な関心事であり、そこではショットノイズが結果の品質に直接影響します。",{"type":21,"tag":18616,"props":18724,"children":18725},{},[],{"type":21,"tag":59,"props":18727,"children":18729},{"id":18728},"_3-quantum-advantage",[18730],{"type":26,"value":18731},"3. Quantum Advantage",{"type":21,"tag":22,"props":18733,"children":18734},{},[18735],{"type":21,"tag":71,"props":18736,"children":18737},{},[18738],{"type":26,"value":18739},"AIとの対応: 10倍の高速化、「人間より優れている」",{"type":21,"tag":22,"props":18741,"children":18742},{},[18743,18745,18750],{"type":26,"value":18744},"「Quantum advantage」とは、量子コンピュータが特定の実世界の問題を、いかなる古典コンピュータよりも ",{"type":21,"tag":51,"props":18746,"children":18747},{},[18748],{"type":26,"value":18749},"速く、あるいはより良く",{"type":26,"value":18751}," 解く瞬間のことです。これはAIベンチマークにおける「超人的な性能」の量子版です。",{"type":21,"tag":22,"props":18753,"children":18754},{},[18755,18757,18761],{"type":26,"value":18756},"IBMは2023年に関連する用語 ",{"type":21,"tag":71,"props":18758,"children":18759},{},[18760],{"type":26,"value":2057},{"type":26,"value":18762}," を導入しました。これは古典的にシミュレートするには複雑すぎるが実用的に有用な回路を表すもので、完全なquantum advantageへのより地に足の着いた前段階です。ハードウェアが成熟するにつれて、両方の用語がプレスリリース、資金調達の発表、規制に関する議論に登場することを予想してください。",{"type":21,"tag":22,"props":18764,"children":18765},{},[18766,18768,18773],{"type":26,"value":18767},"この区別は重要です。quantum advantageは ",{"type":21,"tag":51,"props":18769,"children":18770},{},[18771],{"type":26,"value":18772},"問題とハードウェアの組み合わせ",{"type":26,"value":18774}," の性質であり、量子コンピュータが「より優れている」という包括的な主張ではありません。これを理解することが、情報に基づいた量子の会話を誇大宣伝から分けることになります。",{"type":21,"tag":18616,"props":18776,"children":18777},{},[],{"type":21,"tag":59,"props":18779,"children":18781},{"id":18780},"_4-quantum-utility",[18782],{"type":26,"value":18783},"4. Quantum Utility",{"type":21,"tag":22,"props":18785,"children":18786},{},[18787],{"type":21,"tag":71,"props":18788,"children":18789},{},[18790],{"type":26,"value":18791},"AIとの対応: 「本番AI」対 研究デモ",{"type":21,"tag":22,"props":18793,"children":18794},{},[18795,18797,18801],{"type":26,"value":18796},"quantum advantageと密接に関連する ",{"type":21,"tag":71,"props":18798,"children":18799},{},[18800],{"type":26,"value":2057},{"type":26,"value":18802}," は、具体的には次のことを意味します。すなわち、そのまさに同じ回路の古典的シミュレーションが原理上まだ可能であっても実用的ではない場合に、実際のアプリケーションに有用な結果を生み出す量子回路のことです。",{"type":21,"tag":22,"props":18804,"children":18805},{},[18806,18808,18813],{"type":26,"value":18807},"IBMは2023年の ",{"type":21,"tag":51,"props":18809,"children":18810},{},[18811],{"type":26,"value":18812},"Nature",{"type":26,"value":18814}," 論文で初めてこの用語を使い、自社のEagleプロセッサ上で動作する特定の量子回路が、古典的シミュレーションでは容易に検証できない結果を生み出すことを実証しました。これは完全なquantum advantageよりも達成可能で近い将来のマイルストーンであり、今後数年で最も耳にする可能性が高いものです。",{"type":21,"tag":18616,"props":18816,"children":18817},{},[],{"type":21,"tag":59,"props":18819,"children":18821},{"id":18820},"_5-transpilation",[18822],{"type":26,"value":18823},"5. Transpilation",{"type":21,"tag":22,"props":18825,"children":18826},{},[18827],{"type":21,"tag":71,"props":18828,"children":18829},{},[18830],{"type":26,"value":18831},"AIとの対応: モデルのコンパイル \u002F 量子化",{"type":21,"tag":22,"props":18833,"children":18834},{},[18835,18837,18842],{"type":26,"value":18836},"量子回路が実際のハードウェア上で実行される前に、それは ",{"type":21,"tag":51,"props":18838,"children":18839},{},[18840],{"type":26,"value":18841},"トランスパイル",{"type":26,"value":18843}," されなければなりません。すなわち、抽象的な演算からハードウェアのネイティブなゲートセットへと変換され、量子ビットの演算は物理的な接続性の制約を尊重するように再ルーティングされます。これは、ディープラーニングモデルが特定のチップアーキテクチャ向けにコンパイルされ最適化されなければならないのと類似しています。",{"type":21,"tag":312,"props":18845,"children":18847},{"className":3626,"code":18846,"language":3628,"meta":7,"style":7},"from qiskit.compiler import transpile\n\n# Circuit written in abstract gates\ntranspiled = transpile(circuit, backend=real_qpu, optimization_level=3)\n",[18848],{"type":21,"tag":79,"props":18849,"children":18850},{"__ignoreMap":7},[18851,18872,18879,18887],{"type":21,"tag":322,"props":18852,"children":18853},{"class":324,"line":325},[18854,18858,18863,18867],{"type":21,"tag":322,"props":18855,"children":18856},{"style":3638},[18857],{"type":26,"value":3641},{"type":21,"tag":322,"props":18859,"children":18860},{"style":329},[18861],{"type":26,"value":18862}," qiskit.compiler ",{"type":21,"tag":322,"props":18864,"children":18865},{"style":3638},[18866],{"type":26,"value":3651},{"type":21,"tag":322,"props":18868,"children":18869},{"style":329},[18870],{"type":26,"value":18871}," transpile\n",{"type":21,"tag":322,"props":18873,"children":18874},{"class":324,"line":335},[18875],{"type":21,"tag":322,"props":18876,"children":18877},{"emptyLinePlaceholder":3662},[18878],{"type":26,"value":3665},{"type":21,"tag":322,"props":18880,"children":18881},{"class":324,"line":344},[18882],{"type":21,"tag":322,"props":18883,"children":18884},{"style":3888},[18885],{"type":26,"value":18886},"# Circuit written in abstract gates\n",{"type":21,"tag":322,"props":18888,"children":18889},{"class":324,"line":407},[18890,18895,18899,18904,18909,18913,18918,18922,18926,18930],{"type":21,"tag":322,"props":18891,"children":18892},{"style":329},[18893],{"type":26,"value":18894},"transpiled ",{"type":21,"tag":322,"props":18896,"children":18897},{"style":3638},[18898],{"type":26,"value":3678},{"type":21,"tag":322,"props":18900,"children":18901},{"style":329},[18902],{"type":26,"value":18903}," transpile(circuit, ",{"type":21,"tag":322,"props":18905,"children":18906},{"style":3703},[18907],{"type":26,"value":18908},"backend",{"type":21,"tag":322,"props":18910,"children":18911},{"style":3638},[18912],{"type":26,"value":3678},{"type":21,"tag":322,"props":18914,"children":18915},{"style":329},[18916],{"type":26,"value":18917},"real_qpu, ",{"type":21,"tag":322,"props":18919,"children":18920},{"style":3703},[18921],{"type":26,"value":6712},{"type":21,"tag":322,"props":18923,"children":18924},{"style":3638},[18925],{"type":26,"value":3678},{"type":21,"tag":322,"props":18927,"children":18928},{"style":348},[18929],{"type":26,"value":6948},{"type":21,"tag":322,"props":18931,"children":18932},{"style":329},[18933],{"type":26,"value":3738},{"type":21,"tag":22,"props":18935,"children":18936},{},[18937],{"type":26,"value":18938},"量子が本番環境へと移行するにつれて、「トランスパイルのオーバーヘッド」や「トランスパイルの深さ」が標準的なエンジニアリングの関心事になるでしょう。ちょうど今日、推論のレイテンシやモデルの量子化がそうであるように。",{"type":21,"tag":18616,"props":18940,"children":18941},{},[],{"type":21,"tag":59,"props":18943,"children":18945},{"id":18944},"_6-coherence-time-t1-t2",[18946],{"type":26,"value":18947},"6. Coherence Time (T1 \u002F T2)",{"type":21,"tag":22,"props":18949,"children":18950},{},[18951],{"type":21,"tag":71,"props":18952,"children":18953},{},[18954],{"type":26,"value":18955},"AIとの対応: コンテキストウィンドウ \u002F メモリ制限",{"type":21,"tag":22,"props":18957,"children":18958},{},[18959,18961,18965,18967,18972,18974,18979],{"type":26,"value":18960},"量子ビットは永遠に量子のままではいられません。それは崩壊します。",{"type":21,"tag":51,"props":18962,"children":18963},{},[18964],{"type":26,"value":9369},{"type":26,"value":18966}," と呼ばれるプロセスにおいて、環境ノイズを通じて重ね合わせを失うのです。",{"type":21,"tag":71,"props":18968,"children":18969},{},[18970],{"type":26,"value":18971},"T1時間",{"type":26,"value":18973}," (エネルギー緩和) と ",{"type":21,"tag":71,"props":18975,"children":18976},{},[18977],{"type":26,"value":18978},"T2時間",{"type":26,"value":18980}," (位相コヒーレンス) は、量子ビットがどれだけ長く使用可能な状態を保つかを測定します。",{"type":21,"tag":22,"props":18982,"children":18983},{},[18984,18986,18991],{"type":26,"value":18985},"これはコンテキストウィンドウの量子版です。それは、量子ビットが古典的ノイズになる前に実行できる ",{"type":21,"tag":51,"props":18987,"children":18988},{},[18989],{"type":26,"value":18990},"最大回路深さ",{"type":26,"value":18992}," を定義します。IBMの超伝導量子ビットのT2時間は約100〜300マイクロ秒です。IonQのトラップイオン量子ビットのT2時間は1秒を超えており、これが彼らの回路がはるかに深くなれる理由です。",{"type":21,"tag":200,"props":18994,"children":18995},{},[18996],{"type":21,"tag":22,"props":18997,"children":18998},{},[18999],{"type":26,"value":19000},"「私たちの回路はこのバックエンドには深すぎます。最後のゲートが実行される前にT2に到達してしまうでしょう。」",{"type":21,"tag":22,"props":19002,"children":19003},{},[19004],{"type":26,"value":19005},"開発者は、メモリを予算立てするのと同じように、コヒーレンス時間を予算立てすることを学ぶでしょう。",{"type":21,"tag":18616,"props":19007,"children":19008},{},[],{"type":21,"tag":59,"props":19010,"children":19012},{"id":19011},"_7-decoherence-budget",[19013],{"type":26,"value":19014},"7. Decoherence Budget",{"type":21,"tag":22,"props":19016,"children":19017},{},[19018],{"type":21,"tag":71,"props":19019,"children":19020},{},[19021],{"type":26,"value":19022},"AIとの対応: レイテンシ予算 \u002F 計算予算",{"type":21,"tag":22,"props":19024,"children":19025},{},[19026,19028,19033],{"type":26,"value":19027},"T1\u002FT2の自然な拡張です。",{"type":21,"tag":71,"props":19029,"children":19030},{},[19031],{"type":26,"value":19032},"decoherence budget",{"type":26,"value":19034}," とは、エラーが有用性を超えて蓄積する前に回路が完了するために利用できる「コヒーレンス時間」の総量です。長い回路ほど多くのdecoherence budgetを消費します。",{"type":21,"tag":22,"props":19036,"children":19037},{},[19038],{"type":26,"value":19039},"量子アプリケーションがより複雑になるにつれて、アーキテクトはdecoherence budget内に収まるシステムを設計するようになるでしょう。回路の深さ対結果の忠実度をトレードオフしながら、それは今日、バックエンドエンジニアが応答レイテンシ対計算コストをトレードオフするのと似ています。",{"type":21,"tag":18616,"props":19041,"children":19042},{},[],{"type":21,"tag":59,"props":19044,"children":19046},{"id":19045},"_8-circuit-fidelity",[19047],{"type":26,"value":19048},"8. Circuit Fidelity",{"type":21,"tag":22,"props":19050,"children":19051},{},[19052],{"type":21,"tag":71,"props":19053,"children":19054},{},[19055],{"type":26,"value":19056},"AIとの対応: モデルの精度 \u002F F1スコア",{"type":21,"tag":22,"props":19058,"children":19059},{},[19060,19064],{"type":21,"tag":71,"props":19061,"children":19062},{},[19063],{"type":26,"value":14069},{"type":26,"value":19065}," は、回路によって生成された実際の量子状態が、理想的な理論上の状態にどれだけ近いかを測定します。99%の忠実度を持つ回路は優秀です。90%の回路は精度が求められるアプリケーションには役に立たないかもしれません。",{"type":21,"tag":22,"props":19067,"children":19068},{},[19069],{"type":26,"value":19070},"Fidelityは適用されるすべてのゲートとともに劣化し (各ゲートにはエラー率があります)、回路の深さとともに劣化します (長い回路ほど多くデコヒーレンスを起こします)。それはQPUの結果に対する主要な品質指標になるでしょう。ベンチマーク上のモデル精度の量子版です。",{"type":21,"tag":18616,"props":19072,"children":19073},{},[],{"type":21,"tag":59,"props":19075,"children":19077},{"id":19076},"_9-physical-vs-logical-qubit",[19078],{"type":26,"value":19079},"9. Physical vs. Logical Qubit",{"type":21,"tag":22,"props":19081,"children":19082},{},[19083],{"type":21,"tag":71,"props":19084,"children":19085},{},[19086],{"type":26,"value":19087},"AIとの対応: 生のパラメータ数 対 実効的なモデル容量",{"type":21,"tag":22,"props":19089,"children":19090},{},[19091,19093,19097,19099,19103,19105,19110],{"type":26,"value":19092},"主流になりつつある最も重要な区別の一つです。",{"type":21,"tag":71,"props":19094,"children":19095},{},[19096],{"type":26,"value":9361},{"type":26,"value":19098}," は実際のハードウェア量子ビット (ノイズが多く、エラーを起こしやすい) です。",{"type":21,"tag":71,"props":19100,"children":19101},{},[19102],{"type":26,"value":9390},{"type":26,"value":19104}," は、多数の物理量子ビットにまたがってエンコードされた ",{"type":21,"tag":51,"props":19106,"children":19107},{},[19108],{"type":26,"value":19109},"エラー訂正された",{"type":26,"value":19111}," 量子ビットであり、信頼性は高いですがコストがかかります。",{"type":21,"tag":22,"props":19113,"children":19114},{},[19115,19117,19122],{"type":26,"value":19116},"今日の最良のハードウェア (IBM、IonQ) は物理量子ビットで動作しています。完全な誤り耐性量子コンピューティングには論理量子ビットが必要です。現在の推定では、表面符号によるエラー訂正のために ",{"type":21,"tag":71,"props":19118,"children":19119},{},[19120],{"type":26,"value":19121},"論理量子ビット1個あたり物理量子ビット1,000〜10,000個",{"type":26,"value":19123}," という比率が置かれています。",{"type":21,"tag":22,"props":19125,"children":19126},{},[19127],{"type":26,"value":19128},"「IBMは1,000量子ビットを持っている」と聞くとき、それは物理量子ビットのことです。RSA-2048に対してShorのアルゴリズムを実行できる誤り耐性コンピュータには、数百万個の物理量子ビットが必要になるでしょう。",{"type":21,"tag":22,"props":19130,"children":19131},{},[19132],{"type":26,"value":19133},"この区別を理解することは、量子ハードウェアのマーケティングを技術的現実から分けるために不可欠となるでしょう。",{"type":21,"tag":18616,"props":19135,"children":19136},{},[],{"type":21,"tag":59,"props":19138,"children":19140},{"id":19139},"_10-quantum-safe-post-quantum-cryptography",[19141],{"type":26,"value":19142},"10. Quantum-Safe \u002F Post-Quantum Cryptography",{"type":21,"tag":22,"props":19144,"children":19145},{},[19146],{"type":21,"tag":71,"props":19147,"children":19148},{},[19149],{"type":26,"value":19150},"AIとの対応: 「AI耐性」 (あまり適切ではない) → 実際には「アルゴリズムのためのGDPR」に近い",{"type":21,"tag":22,"props":19152,"children":19153},{},[19154,19156,19161],{"type":26,"value":19155},"これはすでに研究の世界を離れつつあります。",{"type":21,"tag":71,"props":19157,"children":19158},{},[19159],{"type":26,"value":19160},"Post-quantum cryptography",{"type":26,"value":19162}," (PQC) とは、量子コンピュータからの攻撃に耐えるように設計された古典的な暗号アルゴリズムを指します。脅威とは、ShorのアルゴリズムがRSAの基盤となる大きな素数を因数分解でき、今日のほとんどの公開鍵暗号を破ってしまうことです。",{"type":21,"tag":22,"props":19164,"children":19165},{},[19166,19168,19173],{"type":26,"value":19167},"NISTは2024年に最初のポスト量子暗号標準 (CRYSTALS-Kyber、CRYSTALS-Dilithium) を確定しました。",{"type":21,"tag":71,"props":19169,"children":19170},{},[19171],{"type":26,"value":19172},"「Quantum-safe」",{"type":26,"value":19174}," は5年以内に認証およびコンプライアンスの用語となり、セキュリティ監査、調達要件、クラウドプロバイダーのドキュメントに登場するでしょう。",{"type":21,"tag":22,"props":19176,"children":19177},{},[19178,19180,19185],{"type":26,"value":19179},"ほとんどの組織が備えるべき攻撃ベクトルは ",{"type":21,"tag":71,"props":19181,"children":19182},{},[19183],{"type":26,"value":19184},"「harvest now, decrypt later」",{"type":26,"value":19186}," です。すなわち、量子コンピュータが十分に高性能になった時点で復号する意図をもって、今日暗号化されたデータを収集する敵対者のことです。",{"type":21,"tag":18616,"props":19188,"children":19189},{},[],{"type":21,"tag":59,"props":19191,"children":19193},{"id":19192},"_11-hybrid-algorithm-quantum-classical-hybrid",[19194],{"type":26,"value":19195},"11. Hybrid Algorithm \u002F Quantum-Classical Hybrid",{"type":21,"tag":22,"props":19197,"children":19198},{},[19199],{"type":21,"tag":71,"props":19200,"children":19201},{},[19202],{"type":26,"value":19203},"AIとの対応: 「AI支援」ワークフロー、CPU+GPUヘテロジニアスコンピューティング",{"type":21,"tag":22,"props":19205,"children":19206},{},[19207,19209,19214],{"type":26,"value":19208},"ほぼすべての近未来の量子アプリケーションは ",{"type":21,"tag":71,"props":19210,"children":19211},{},[19212],{"type":26,"value":19213},"hybrid",{"type":26,"value":19215}," です。古典的なオプティマイザが量子回路を繰り返し呼び出し、古典コンピュータが苦手とする部分 (状態準備、干渉) には量子ハードウェアを使い、一方で古典コンピュータが最適化ループを処理します。",{"type":21,"tag":22,"props":19217,"children":19218},{},[19219],{"type":26,"value":19220},"VQEとQAOAは典型的なハイブリッドアルゴリズムです。量子プログラミングフレームワークが成熟するにつれて、「ハイブリッドワークフローのオーケストレーション」 — 古典と量子の計算間の受け渡しを管理すること — が標準的なソフトウェアエンジニアリングの問題になるでしょう。それを中心に構築されたツール、フレームワーク、職種を予想してください。",{"type":21,"tag":18616,"props":19222,"children":19223},{},[],{"type":21,"tag":59,"props":19225,"children":19227},{"id":19226},"_12-variational-circuit-parameterized-quantum-circuit",[19228],{"type":26,"value":19229},"12. Variational Circuit (Parameterized Quantum Circuit)",{"type":21,"tag":22,"props":19231,"children":19232},{},[19233],{"type":21,"tag":71,"props":19234,"children":19235},{},[19236],{"type":26,"value":19237},"AIとの対応: ニューラルネットワーク (訓練可能な重み)",{"type":21,"tag":22,"props":19239,"children":19240},{},[19241,19245,19247,19252],{"type":21,"tag":71,"props":19242,"children":19243},{},[19244],{"type":26,"value":14274},{"type":26,"value":19246}," または ",{"type":21,"tag":71,"props":19248,"children":19249},{},[19250],{"type":26,"value":19251},"parameterized quantum circuit (PQC)",{"type":26,"value":19253}," は、調整可能な回転角を持つ量子回路であり、本質的には量子ニューラルネットワークの層です。パラメータは、コスト関数を最小化するために古典的なオプティマイザによって調整されます。",{"type":21,"tag":22,"props":19255,"children":19256},{},[19257],{"type":26,"value":19258},"VQEは変分回路をエネルギー推定器として使用します。QAOAはそれを最適化問題をエンコードするために使用します。量子機械学習が発展するにつれて、「量子回路を訓練する」は「ニューラルネットワークを訓練する」と同じくらい自然な言い回しになるでしょう。",{"type":21,"tag":18616,"props":19260,"children":19261},{},[],{"type":21,"tag":59,"props":19263,"children":19265},{"id":19264},"_13-quantum-volume-qv",[19266],{"type":26,"value":19267},"13. Quantum Volume (QV)",{"type":21,"tag":22,"props":19269,"children":19270},{},[19271],{"type":21,"tag":71,"props":19272,"children":19273},{},[19274],{"type":26,"value":19275},"AIとの対応: FLOPS、ベンチマークスコア (MLPerf)",{"type":21,"tag":22,"props":19277,"children":19278},{},[19279,19283],{"type":21,"tag":71,"props":19280,"children":19281},{},[19282],{"type":26,"value":9465},{"type":26,"value":19284}," はIBMによって導入された単一の数値ベンチマークで、QPUの性能を総合的に捉えるものです。量子ビット数、ゲート忠実度、接続性、実行可能な最大回路深さを同時に考慮します。",{"type":21,"tag":22,"props":19286,"children":19287},{},[19288],{"type":26,"value":19289},"QV 128 (= 2^7、7量子ビット相当の回路) を持つQPUは、たとえQVの低いマシンの方が生の量子ビット数が多くても、QV 64を持つものを上回ります。これは量子の世界の「MLPerfスコア」です。ベンダーが宣伝し、開発者がハードウェアを比較するために使う数値です。",{"type":21,"tag":18616,"props":19291,"children":19292},{},[],{"type":21,"tag":59,"props":19294,"children":19296},{"id":19295},"_14-quantum-job-quantum-queue",[19297],{"type":26,"value":19298},"14. Quantum Job \u002F Quantum Queue",{"type":21,"tag":22,"props":19300,"children":19301},{},[19302],{"type":21,"tag":71,"props":19303,"children":19304},{},[19305],{"type":26,"value":19306},"AIとの対応: バッチ推論ジョブ、GPUキュー",{"type":21,"tag":22,"props":19308,"children":19309},{},[19310,19312,19317,19319,19324],{"type":26,"value":19311},"実際のQPUに回路を投入するとき、あなたは ",{"type":21,"tag":71,"props":19313,"children":19314},{},[19315],{"type":26,"value":19316},"quantum queue",{"type":26,"value":19318}," に入る ",{"type":21,"tag":71,"props":19320,"children":19321},{},[19322],{"type":26,"value":19323},"quantum job",{"type":26,"value":19325}," を投入します。これは共有ハードウェアへのアクセスを管理するスケジューリングシステムです。IBM Quantumの無料枠のジョブは、有料枠の顧客の後ろで、しばしば数分から数時間キューで待機します。",{"type":21,"tag":22,"props":19327,"children":19328},{},[19329],{"type":26,"value":19330},"「ジョブスループット」「キューの深さ」「ジョブ優先度」が、量子インフラチームにとっての運用指標になるでしょう。これは今日GPUクラスタのジョブキューを管理するのと精神的に同一です。",{"type":21,"tag":18616,"props":19332,"children":19333},{},[],{"type":21,"tag":59,"props":19335,"children":19337},{"id":19336},"_15-error-mitigation-vs-error-correction",[19338],{"type":26,"value":19339},"15. Error Mitigation vs. Error Correction",{"type":21,"tag":22,"props":19341,"children":19342},{},[19343],{"type":21,"tag":71,"props":19344,"children":19345},{},[19346],{"type":26,"value":19347},"AIとの対応: 正則化 (mitigation) 対 ハードウェアECC (correction)",{"type":21,"tag":22,"props":19349,"children":19350},{},[19351],{"type":26,"value":19352},"この区別は重要であり、量子が主流になれば絶えず誤用されるでしょう。",{"type":21,"tag":22,"props":19354,"children":19355},{},[19356,19361],{"type":21,"tag":71,"props":19357,"children":19358},{},[19359],{"type":26,"value":19360},"Error correction",{"type":26,"value":19362}," — 冗長性をもって論理量子ビットをエンコードし、エラーを検出して修正できるようにすること。約1,000:1の物理対論理量子ビットのオーバーヘッドを必要とします。まだ大規模には利用できません。",{"type":21,"tag":22,"props":19364,"children":19365},{},[19366,19370,19372,19377],{"type":21,"tag":71,"props":19367,"children":19368},{},[19369],{"type":26,"value":4487},{"type":26,"value":19371}," — ノイズを物理的に修正 ",{"type":21,"tag":51,"props":19373,"children":19374},{},[19375],{"type":26,"value":19376},"することなく",{"type":26,"value":19378}," 、その影響を統計的に低減する古典的な後処理技術。ゼロノイズ外挿、確率的エラーキャンセル、測定エラー緩和など。NISQハードウェア上で、今すぐ利用可能です。",{"type":21,"tag":22,"props":19380,"children":19381},{},[19382],{"type":26,"value":19383},"ベンダーが自社のQPUが「エラーを処理する」と主張するときは、常にどちらを意味しているのか尋ねてください。",{"type":21,"tag":18616,"props":19385,"children":19386},{},[],{"type":21,"tag":59,"props":19388,"children":19390},{"id":19389},"_16-quantum-developer-quantum-engineer",[19391],{"type":26,"value":19392},"16. Quantum Developer \u002F Quantum Engineer",{"type":21,"tag":22,"props":19394,"children":19395},{},[19396],{"type":21,"tag":71,"props":19397,"children":19398},{},[19399],{"type":26,"value":19400},"AIとの対応: MLエンジニア、AIエンジニア、プロンプトエンジニア",{"type":21,"tag":22,"props":19402,"children":19403},{},[19404,19406,19411],{"type":26,"value":19405},"おそらく最も重大な新しい用語は ",{"type":21,"tag":71,"props":19407,"children":19408},{},[19409],{"type":26,"value":19410},"quantum developer",{"type":26,"value":19412}," です。「AIエンジニア」の役割が2023年頃に結晶化したのと同じように — 従来のソフトウェアエンジニアリングともML研究とも異なるものとして — 「quantum developer」の役割は特定のスキルセットを中心に生まれつつあります。",{"type":21,"tag":3556,"props":19414,"children":19415},{},[19416,19421,19426,19431],{"type":21,"tag":3560,"props":19417,"children":19418},{},[19419],{"type":26,"value":19420},"Qiskit、Cirq、HLQuantumのようなフレームワークで回路を書くこと",{"type":21,"tag":3560,"props":19422,"children":19423},{},[19424],{"type":26,"value":19425},"ハードウェアの制約 (接続性、ノイズ、コヒーレンス) を理解すること",{"type":21,"tag":3560,"props":19427,"children":19428},{},[19429],{"type":26,"value":19430},"ハイブリッドな古典-量子ワークフローを設計すること",{"type":21,"tag":3560,"props":19432,"children":19433},{},[19434],{"type":26,"value":19435},"確率的な結果と忠実度指標を解釈すること",{"type":21,"tag":22,"props":19437,"children":19438},{},[19439,19445],{"type":21,"tag":29,"props":19440,"children":19442},{"href":19441},"\u002Fhlquantum",[19443],{"type":26,"value":19444},"HLQuantum",{"type":26,"value":19446}," のようなツール — すべてのバックエンドにまたがる統一されたAPIを提供する — は、開発者が各SDKの癖に深い専門知識を持たずとも移植可能な量子コードを書けるようにすることで、これを加速しています。2028年の「quantum developer」は、今日のWeb開発者がフレームワークを使うのと同じように抽象化レイヤーを使い、各バックエンドごとに生の回路を書くことはおそらくないでしょう。",{"type":21,"tag":18616,"props":19448,"children":19449},{},[],{"type":21,"tag":59,"props":19451,"children":19453},{"id":19452},"パターン",[19454],{"type":26,"value":19452},{"type":21,"tag":22,"props":19456,"children":19457},{},[19458],{"type":26,"value":19459},"これら16の用語を見渡すと、あるパターンが浮かび上がります。あらゆる主要なプラットフォームの転換は次のものを生み出します。",{"type":21,"tag":7444,"props":19461,"children":19462},{},[19463,19473,19483,19493,19503,19513,19523],{"type":21,"tag":3560,"props":19464,"children":19465},{},[19466,19471],{"type":21,"tag":71,"props":19467,"children":19468},{},[19469],{"type":26,"value":19470},"新しい計算の単位",{"type":26,"value":19472}," — GPU → QPU、tokens → shots",{"type":21,"tag":3560,"props":19474,"children":19475},{},[19476,19481],{"type":21,"tag":71,"props":19477,"children":19478},{},[19479],{"type":26,"value":19480},"新しい資源の制約",{"type":26,"value":19482}," — VRAM、コンテキストウィンドウ → コヒーレンス時間、shot budget",{"type":21,"tag":3560,"props":19484,"children":19485},{},[19486,19491],{"type":21,"tag":71,"props":19487,"children":19488},{},[19489],{"type":26,"value":19490},"新しい品質指標",{"type":26,"value":19492}," — 精度、F1 → 忠実度、quantum volume",{"type":21,"tag":3560,"props":19494,"children":19495},{},[19496,19501],{"type":21,"tag":71,"props":19497,"children":19498},{},[19499],{"type":26,"value":19500},"新しいコンパイルステップ",{"type":26,"value":19502}," — モデルの量子化 → トランスパイル",{"type":21,"tag":3560,"props":19504,"children":19505},{},[19506,19511],{"type":21,"tag":71,"props":19507,"children":19508},{},[19509],{"type":26,"value":19510},"新しいハイブリッドアーキテクチャ",{"type":26,"value":19512}," — CPU+GPU → 古典+量子",{"type":21,"tag":3560,"props":19514,"children":19515},{},[19516,19521],{"type":21,"tag":71,"props":19517,"children":19518},{},[19519],{"type":26,"value":19520},"新しいセキュリティの領域",{"type":26,"value":19522}," — 敵対的ML → ポスト量子暗号",{"type":21,"tag":3560,"props":19524,"children":19525},{},[19526,19531],{"type":21,"tag":71,"props":19527,"children":19528},{},[19529],{"type":26,"value":19530},"新しい職種",{"type":26,"value":19532}," — MLエンジニア → 量子エンジニア",{"type":21,"tag":22,"props":19534,"children":19535},{},[19536],{"type":26,"value":19537},"量子コンピューティングの語彙は単なる専門用語ではありません。それは実際のエンジニアリングの制約、実際のトレードオフ、実際の機会に直接対応しています。主流の波が到来する前にそれを学ぶことは、初期のクラウドエンジニアが、ほとんどの開発者よりも先に「レイテンシ」や「スループット」を気にかけ始めたときに賭けたのと同じ賭けです。",{"type":21,"tag":22,"props":19539,"children":19540},{},[19541],{"type":26,"value":19542},"始めるのに最も良い時期は5年前でした。次に良い時期は今です。ハードウェアがまだ無料でアクセスでき、学習曲線が最も低いうちに。",{"type":21,"tag":22,"props":19544,"children":19545},{},[19546,19551,19552],{"type":21,"tag":71,"props":19547,"children":19548},{},[19549],{"type":26,"value":19550},"ここから始めましょう:",{"type":26,"value":1901},{"type":21,"tag":29,"props":19553,"children":19554},{"href":1005},[19555],{"type":26,"value":19556},"初めての量子回路を無料で実行する →",{"type":21,"tag":703,"props":19558,"children":19559},{},[19560],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":19562},[19563,19564,19565,19566,19567,19568,19569,19570,19571,19572,19573,19574,19575,19576,19577,19578,19579],{"id":18621,"depth":335,"text":18624},{"id":18674,"depth":335,"text":18677},{"id":18728,"depth":335,"text":18731},{"id":18780,"depth":335,"text":18783},{"id":18820,"depth":335,"text":18823},{"id":18944,"depth":335,"text":18947},{"id":19011,"depth":335,"text":19014},{"id":19045,"depth":335,"text":19048},{"id":19076,"depth":335,"text":19079},{"id":19139,"depth":335,"text":19142},{"id":19192,"depth":335,"text":19195},{"id":19226,"depth":335,"text":19229},{"id":19264,"depth":335,"text":19267},{"id":19295,"depth":335,"text":19298},{"id":19336,"depth":335,"text":19339},{"id":19389,"depth":335,"text":19392},{"id":19452,"depth":335,"text":19452},"content:ja:blog:quantum-computing-new-vocabulary.md","ja\u002Fblog\u002Fquantum-computing-new-vocabulary.md","ja\u002Fblog\u002Fquantum-computing-new-vocabulary",{"_path":19584,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":19585,"description":19586,"date":19587,"author":11,"tags":19588,"readingTime":16,"body":19589,"_type":715,"_id":20599,"_source":717,"_file":20600,"_stem":20601,"_extension":720},"\u002Fja\u002Fblog\u002Fquantum-use-cases-2026","2026年に量子コンピューターが実際にできること","どの量子コンピューティング応用が今すぐ本番投入できるのか、どれが近い将来のものなのか、どれが依然として長期的な研究目標のままなのかを現実的に見つめ、本物の進歩を誇大宣伝から切り分ける。","2026-03-12",[1033,15,7585],{"type":18,"children":19590,"toc":20579},[19591,19596,19601,19606,19611,19644,19649,19652,19658,19664,19674,19684,19694,19704,19714,19879,19882,19888,19897,19906,19924,19934,19943,19946,19952,19961,19978,19987,20004,20007,20013,20022,20031,20048,20058,20061,20067,20072,20081,20090,20100,20110,20113,20119,20144,20153,20162,20165,20171,20180,20189,20198,20207,20210,20216,20222,20231,20241,20251,20268,20271,20277,20286,20295,20304,20307,20312,20321,20330,20339,20342,20347,20513,20532,20544,20549,20575],{"type":21,"tag":22,"props":19592,"children":19593},{},[19594],{"type":26,"value":19595},"「量子コンピューターは創薬に革命を起こし、あらゆる暗号を破り、気候変動を解決する」。そんな売り文句を聞いたことがあるでしょう。では正直な答えをお伝えします。2026年、今日存在するハードウェア上で、量子コンピューターは実際に何ができるのでしょうか。",{"type":21,"tag":22,"props":19597,"children":19598},{},[19599],{"type":26,"value":19600},"正直な答えは微妙なニュアンスを含んでいます。そして、誇大宣伝や懐疑論のどちらが示唆するよりも興味深いものです。",{"type":21,"tag":59,"props":19602,"children":19604},{"id":19603},"量子の実用度をどう考えるか",[19605],{"type":26,"value":19603},{"type":21,"tag":22,"props":19607,"children":19608},{},[19609],{"type":26,"value":19610},"量子応用は3つのカテゴリーに分けられます。",{"type":21,"tag":3556,"props":19612,"children":19613},{},[19614,19624,19634],{"type":21,"tag":3560,"props":19615,"children":19616},{},[19617,19622],{"type":21,"tag":71,"props":19618,"children":19619},{},[19620],{"type":26,"value":19621},"NISQ時代の応用(現在):",{"type":26,"value":19623}," 限られた回路深度を持つ、今日のノイズの多い50〜1000量子ビットのハードウェア上で動作するアルゴリズム。結果は近似的である場合があります。価値は探索的またはデモンストレーション的なものです。",{"type":21,"tag":3560,"props":19625,"children":19626},{},[19627,19632],{"type":21,"tag":71,"props":19628,"children":19629},{},[19630],{"type":26,"value":19631},"近い将来の応用(2〜7年):",{"type":26,"value":19633}," 適度な誤り訂正(数百から数千の論理量子ビット)を必要とします。特定の問題インスタンスに対する実用的な高速化が現れ始めます。",{"type":21,"tag":3560,"props":19635,"children":19636},{},[19637,19642],{"type":21,"tag":71,"props":19638,"children":19639},{},[19640],{"type":26,"value":19641},"長期的な応用(7〜20年以上):",{"type":26,"value":19643}," 数百万の物理量子ビットと完全なフォールトトレランスを必要とします。変革的なインパクトをもたらします。",{"type":21,"tag":22,"props":19645,"children":19646},{},[19647],{"type":26,"value":19648},"誇大宣伝と現実の間のギャップは、主にこれら3つのカテゴリーを混同することから生じます。",{"type":21,"tag":18616,"props":19650,"children":19651},{},[],{"type":21,"tag":59,"props":19653,"children":19655},{"id":19654},"今できることnisq時代",[19656],{"type":26,"value":19657},"今できること(NISQ時代)",{"type":21,"tag":17800,"props":19659,"children":19661},{"id":19660},"量子化学シミュレーション-限られた分子",[19662],{"type":26,"value":19663},"量子化学シミュレーション — 限られた分子",{"type":21,"tag":22,"props":19665,"children":19666},{},[19667,19672],{"type":21,"tag":71,"props":19668,"children":19669},{},[19670],{"type":26,"value":19671},"何を:",{"type":26,"value":19673}," VQEを用いた小分子の基底状態エネルギーのシミュレーション。",{"type":21,"tag":22,"props":19675,"children":19676},{},[19677,19682],{"type":21,"tag":71,"props":19678,"children":19679},{},[19680],{"type":26,"value":19681},"実証済み:",{"type":26,"value":19683}," IBMなどが、実際のQPU上でH₂、LiH、BeH₂、および小さな炭化水素の基底状態エネルギーを計算しました。4〜12量子ビットの問題では、結果が古典的手法と一致しています。",{"type":21,"tag":22,"props":19685,"children":19686},{},[19687,19692],{"type":21,"tag":71,"props":19688,"children":19689},{},[19690],{"type":26,"value":19691},"現在の実用的価値:",{"type":26,"value":19693}," 研究とベンチマーク。古典コンピューターでも正確にシミュレートできる分子については優位性はありませんが、手法の検証は進んでいます。",{"type":21,"tag":22,"props":19695,"children":19696},{},[19697,19702],{"type":21,"tag":71,"props":19698,"children":19699},{},[19700],{"type":26,"value":19701},"ボトルネック:",{"type":26,"value":19703}," 実用的な製薬上の関連性を持つ分子(50原子以上)へのスケールアップには数百の論理量子ビットが必要です。現在のノイズの多いハードウェアでは不可能です。",{"type":21,"tag":22,"props":19705,"children":19706},{},[19707,19712],{"type":21,"tag":71,"props":19708,"children":19709},{},[19710],{"type":26,"value":19711},"取り組んでいるのは:",{"type":26,"value":19713}," IBM Quantum、IonQ(創薬パートナーシップ向け)、Quantinuum。",{"type":21,"tag":312,"props":19715,"children":19717},{"className":3626,"code":19716,"language":3628,"meta":7,"style":7},"import hlquantum as hlq\n\n# VQE for H2 — works today, results match classical\nresult = hlq.algorithms.vqe(molecule='H2', basis='sto-3g', backend='qiskit')\nprint(f\"Ground state: {result.energy:.4f} Ha\")  # -1.1372 Ha\n",[19718],{"type":21,"tag":79,"props":19719,"children":19720},{"__ignoreMap":7},[19721,19742,19749,19757,19827],{"type":21,"tag":322,"props":19722,"children":19723},{"class":324,"line":325},[19724,19728,19733,19737],{"type":21,"tag":322,"props":19725,"children":19726},{"style":3638},[19727],{"type":26,"value":3651},{"type":21,"tag":322,"props":19729,"children":19730},{"style":329},[19731],{"type":26,"value":19732}," hlquantum ",{"type":21,"tag":322,"props":19734,"children":19735},{"style":3638},[19736],{"type":26,"value":4715},{"type":21,"tag":322,"props":19738,"children":19739},{"style":329},[19740],{"type":26,"value":19741}," hlq\n",{"type":21,"tag":322,"props":19743,"children":19744},{"class":324,"line":335},[19745],{"type":21,"tag":322,"props":19746,"children":19747},{"emptyLinePlaceholder":3662},[19748],{"type":26,"value":3665},{"type":21,"tag":322,"props":19750,"children":19751},{"class":324,"line":344},[19752],{"type":21,"tag":322,"props":19753,"children":19754},{"style":3888},[19755],{"type":26,"value":19756},"# VQE for H2 — works today, results match classical\n",{"type":21,"tag":322,"props":19758,"children":19759},{"class":324,"line":407},[19760,19765,19769,19774,19779,19783,19788,19792,19797,19801,19806,19810,19814,19818,19823],{"type":21,"tag":322,"props":19761,"children":19762},{"style":329},[19763],{"type":26,"value":19764},"result ",{"type":21,"tag":322,"props":19766,"children":19767},{"style":3638},[19768],{"type":26,"value":3678},{"type":21,"tag":322,"props":19770,"children":19771},{"style":329},[19772],{"type":26,"value":19773}," hlq.algorithms.vqe(",{"type":21,"tag":322,"props":19775,"children":19776},{"style":3703},[19777],{"type":26,"value":19778},"molecule",{"type":21,"tag":322,"props":19780,"children":19781},{"style":3638},[19782],{"type":26,"value":3678},{"type":21,"tag":322,"props":19784,"children":19785},{"style":359},[19786],{"type":26,"value":19787},"'H2'",{"type":21,"tag":322,"props":19789,"children":19790},{"style":329},[19791],{"type":26,"value":367},{"type":21,"tag":322,"props":19793,"children":19794},{"style":3703},[19795],{"type":26,"value":19796},"basis",{"type":21,"tag":322,"props":19798,"children":19799},{"style":3638},[19800],{"type":26,"value":3678},{"type":21,"tag":322,"props":19802,"children":19803},{"style":359},[19804],{"type":26,"value":19805},"'sto-3g'",{"type":21,"tag":322,"props":19807,"children":19808},{"style":329},[19809],{"type":26,"value":367},{"type":21,"tag":322,"props":19811,"children":19812},{"style":3703},[19813],{"type":26,"value":18908},{"type":21,"tag":322,"props":19815,"children":19816},{"style":3638},[19817],{"type":26,"value":3678},{"type":21,"tag":322,"props":19819,"children":19820},{"style":359},[19821],{"type":26,"value":19822},"'qiskit'",{"type":21,"tag":322,"props":19824,"children":19825},{"style":329},[19826],{"type":26,"value":3738},{"type":21,"tag":322,"props":19828,"children":19829},{"class":324,"line":431},[19830,19834,19838,19842,19847,19851,19856,19860,19864,19869,19874],{"type":21,"tag":322,"props":19831,"children":19832},{"style":348},[19833],{"type":26,"value":3753},{"type":21,"tag":322,"props":19835,"children":19836},{"style":329},[19837],{"type":26,"value":3758},{"type":21,"tag":322,"props":19839,"children":19840},{"style":3638},[19841],{"type":26,"value":3763},{"type":21,"tag":322,"props":19843,"children":19844},{"style":359},[19845],{"type":26,"value":19846},"\"Ground state: ",{"type":21,"tag":322,"props":19848,"children":19849},{"style":348},[19850],{"type":26,"value":3773},{"type":21,"tag":322,"props":19852,"children":19853},{"style":329},[19854],{"type":26,"value":19855},"result.energy",{"type":21,"tag":322,"props":19857,"children":19858},{"style":3638},[19859],{"type":26,"value":4085},{"type":21,"tag":322,"props":19861,"children":19862},{"style":348},[19863],{"type":26,"value":3783},{"type":21,"tag":322,"props":19865,"children":19866},{"style":359},[19867],{"type":26,"value":19868}," Ha\"",{"type":21,"tag":322,"props":19870,"children":19871},{"style":329},[19872],{"type":26,"value":19873},")  ",{"type":21,"tag":322,"props":19875,"children":19876},{"style":3888},[19877],{"type":26,"value":19878},"# -1.1372 Ha\n",{"type":21,"tag":18616,"props":19880,"children":19881},{},[],{"type":21,"tag":17800,"props":19883,"children":19885},{"id":19884},"量子機械学習-概念実証",[19886],{"type":26,"value":19887},"量子機械学習 — 概念実証",{"type":21,"tag":22,"props":19889,"children":19890},{},[19891,19895],{"type":21,"tag":71,"props":19892,"children":19893},{},[19894],{"type":26,"value":19671},{"type":26,"value":19896}," 分類のための量子ニューラルネットワーク(QNN)と量子カーネル法。",{"type":21,"tag":22,"props":19898,"children":19899},{},[19900,19904],{"type":21,"tag":71,"props":19901,"children":19902},{},[19903],{"type":26,"value":19681},{"type":26,"value":19905}," 量子カーネルはおもちゃのデータセット上で機能することが示されています。実用的なデータセットに対して、古典的な機械学習に対する量子優位性は実証されていません。",{"type":21,"tag":22,"props":19907,"children":19908},{},[19909,19913,19915,19922],{"type":21,"tag":71,"props":19910,"children":19911},{},[19912],{"type":26,"value":19691},{"type":26,"value":19914}," 研究。「",{"type":21,"tag":29,"props":19916,"children":19919},{"href":19917,"rel":19918},"https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.01aborting",[44],[19920],{"type":26,"value":19921},"ノーフリーランチ",{"type":26,"value":19923},"」問題が当てはまります。QMLにおける量子優位性は、もし存在するとすれば、特定の量子構造を持つ特定のデータに対するものになるでしょう。",{"type":21,"tag":22,"props":19925,"children":19926},{},[19927,19932],{"type":21,"tag":71,"props":19928,"children":19929},{},[19930],{"type":26,"value":19931},"現れつつある現実的なユースケース:",{"type":26,"value":19933}," 量子物理実験から生じるデータセット向けの量子カーネルは、本質的な量子構造を持つ可能性があり、量子機械学習が古典的手法よりもうまくそれを活用できるかもしれません。",{"type":21,"tag":22,"props":19935,"children":19936},{},[19937,19941],{"type":21,"tag":71,"props":19938,"children":19939},{},[19940],{"type":26,"value":19711},{"type":26,"value":19942}," PennyLane\u002FXanadu(QML重視)、IBM、Google(Quantum AI)。",{"type":21,"tag":18616,"props":19944,"children":19945},{},[],{"type":21,"tag":17800,"props":19947,"children":19949},{"id":19948},"最適化のベンチマークqaoa",[19950],{"type":26,"value":19951},"最適化のベンチマーク(QAOA)",{"type":21,"tag":22,"props":19953,"children":19954},{},[19955,19959],{"type":21,"tag":71,"props":19956,"children":19957},{},[19958],{"type":26,"value":19671},{"type":26,"value":19960}," Max-Cut、ポートフォリオ最適化、車両ルーティングに適用されるQAOA。",{"type":21,"tag":22,"props":19962,"children":19963},{},[19964,19968,19970,19976],{"type":21,"tag":71,"props":19965,"children":19966},{},[19967],{"type":26,"value":19681},{"type":26,"value":19969}," QAOA回路は実際のハードウェア上で実行されてきました。",{"type":21,"tag":79,"props":19971,"children":19973},{"className":19972},[],[19974],{"type":26,"value":19975},"p=1",{"type":26,"value":19977},"では、Max-Cutにおいて約0.75の近似比がNISQデバイス上で達成可能です。",{"type":21,"tag":22,"props":19979,"children":19980},{},[19981,19985],{"type":21,"tag":71,"props":19982,"children":19983},{},[19984],{"type":26,"value":19691},{"type":26,"value":19986}," 現実的な問題規模では古典ソルバー(シミュレーテッドアニーリング、GUROBI)に対して競争力はありません。研究とハードウェアのベンチマークには価値があります。",{"type":21,"tag":22,"props":19988,"children":19989},{},[19990,19995,19997,20002],{"type":21,"tag":71,"props":19991,"children":19992},{},[19993],{"type":26,"value":19994},"それが重要になるとき:",{"type":26,"value":19996}," 大きな優位性を得られるほど",{"type":21,"tag":79,"props":19998,"children":20000},{"className":19999},[],[20001],{"type":26,"value":22},{"type":26,"value":20003},"が大きくなると、回路がNISQデバイスには深すぎます。誤り訂正が必要です。",{"type":21,"tag":18616,"props":20005,"children":20006},{},[],{"type":21,"tag":17800,"props":20008,"children":20010},{"id":20009},"ランダム回路サンプリング量子超越性の主張",[20011],{"type":26,"value":20012},"ランダム回路サンプリング(量子超越性の主張)",{"type":21,"tag":22,"props":20014,"children":20015},{},[20016,20020],{"type":21,"tag":71,"props":20017,"children":20018},{},[20019],{"type":26,"value":19671},{"type":26,"value":20021}," ランダム量子回路からのサンプリング。",{"type":21,"tag":22,"props":20023,"children":20024},{},[20025,20029],{"type":21,"tag":71,"props":20026,"children":20027},{},[20028],{"type":26,"value":19681},{"type":26,"value":20030}," Google(2019年、Sycamore)、IBM(2023年、Eagle)、中国のグループ(2020年、2021年)。",{"type":21,"tag":22,"props":20032,"children":20033},{},[20034,20039,20041,20046],{"type":21,"tag":71,"props":20035,"children":20036},{},[20037],{"type":26,"value":20038},"実用的価値:",{"type":26,"value":20040}," 現在のところなし。ランダム回路サンプリングには既知の応用がありません。これは、量子デバイスが古典コンピューターにはシミュレートしにくい",{"type":21,"tag":51,"props":20042,"children":20043},{},[20044],{"type":26,"value":20045},"何か",{"type":26,"value":20047},"を行っていることを示すハードウェアのベンチマークです。",{"type":21,"tag":22,"props":20049,"children":20050},{},[20051,20056],{"type":21,"tag":71,"props":20052,"children":20053},{},[20054],{"type":26,"value":20055},"それが重要な理由:",{"type":26,"value":20057}," 量子ハードウェアが正しく動作していること、そして特定の回路サイズにおいて古典的シミュレーションが手に負えなくなりつつあることを検証します。",{"type":21,"tag":18616,"props":20059,"children":20060},{},[],{"type":21,"tag":59,"props":20062,"children":20064},{"id":20063},"近い将来のこと27年",[20065],{"type":26,"value":20066},"近い将来のこと(2〜7年)",{"type":21,"tag":17800,"props":20068,"children":20070},{"id":20069},"量子支援による創薬",[20071],{"type":26,"value":20069},{"type":21,"tag":22,"props":20073,"children":20074},{},[20075,20079],{"type":21,"tag":71,"props":20076,"children":20077},{},[20078],{"type":26,"value":19671},{"type":26,"value":20080}," 量子シミュレーションを用いてタンパク質-リガンド結合エネルギーをモデル化し、どの候補薬が合成する価値があるかを予測します。",{"type":21,"tag":22,"props":20082,"children":20083},{},[20084,20088],{"type":21,"tag":71,"props":20085,"children":20086},{},[20087],{"type":26,"value":20055},{"type":26,"value":20089}," 創薬には承認薬1つあたり約25億ドルのコストがかかります。より優れた計算的スクリーニングによって失敗する臨床試験を減らすことは、非常に大きな価値があります。",{"type":21,"tag":22,"props":20091,"children":20092},{},[20093,20098],{"type":21,"tag":71,"props":20094,"children":20095},{},[20096],{"type":26,"value":20097},"現状:",{"type":26,"value":20099}," IBMと製薬パートナー(Pfizer、AstraZeneca、JSR)は、量子ハードウェア上で最大約50量子ビットの分子に対してVQEを実行しています。まだデモンストレーション段階であり、商業的に関連する分子については古典的手法の方が依然として正確です。",{"type":21,"tag":22,"props":20101,"children":20102},{},[20103,20108],{"type":21,"tag":71,"props":20104,"children":20105},{},[20106],{"type":26,"value":20107},"それが現実になるとき:",{"type":26,"value":20109}," 論理量子ビット数が約1,000〜5,000に達したとき。そうすれば、主要な分子クラスにおいて量子シミュレーションが古典を上回ることができます。現在の推定では5〜10年。",{"type":21,"tag":18616,"props":20111,"children":20112},{},[],{"type":21,"tag":17800,"props":20114,"children":20116},{"id":20115},"量子金融モンテカルロの高速化",[20117],{"type":26,"value":20118},"量子金融:モンテカルロの高速化",{"type":21,"tag":22,"props":20120,"children":20121},{},[20122,20126,20128,20134,20136,20142],{"type":21,"tag":71,"props":20123,"children":20124},{},[20125],{"type":26,"value":19671},{"type":26,"value":20127}," 量子振幅推定は理論的にモンテカルロ積分を",{"type":21,"tag":79,"props":20129,"children":20131},{"className":20130},[],[20132],{"type":26,"value":20133},"O(1\u002Fε²)",{"type":26,"value":20135},"から",{"type":21,"tag":79,"props":20137,"children":20139},{"className":20138},[],[20140],{"type":26,"value":20141},"O(1\u002Fε)",{"type":26,"value":20143},"へと高速化できます。これはオプション価格計算やリスクシミュレーションに極めて重要な二次的高速化です。",{"type":21,"tag":22,"props":20145,"children":20146},{},[20147,20151],{"type":21,"tag":71,"props":20148,"children":20149},{},[20150],{"type":26,"value":20097},{"type":26,"value":20152}," Goldman Sachs、JPMorgan、BBVAが積極的にこれを研究しています。小規模なデモンストレーションが行われてきました。まだ本番規模ではありません。",{"type":21,"tag":22,"props":20154,"children":20155},{},[20156,20160],{"type":21,"tag":71,"props":20157,"children":20158},{},[20159],{"type":26,"value":20107},{"type":26,"value":20161}," 適度なフォールトトレランス(数千の論理量子ビット)が必要です。実用的な金融面での高速化までの推定は5〜15年の幅があります。",{"type":21,"tag":18616,"props":20163,"children":20164},{},[],{"type":21,"tag":17800,"props":20166,"children":20168},{"id":20167},"材料科学電池と太陽電池の設計",[20169],{"type":26,"value":20170},"材料科学:電池と太陽電池の設計",{"type":21,"tag":22,"props":20172,"children":20173},{},[20174,20178],{"type":21,"tag":71,"props":20175,"children":20176},{},[20177],{"type":26,"value":19671},{"type":26,"value":20179}," リチウム空気電池、窒素固定触媒(ハーバー・ボッシュ法を置き換えるもの)、光起電力材料の候補材料の量子シミュレーション。",{"type":21,"tag":22,"props":20181,"children":20182},{},[20183,20187],{"type":21,"tag":71,"props":20184,"children":20185},{},[20186],{"type":26,"value":20055},{"type":26,"value":20188}," より優れた電池 = EV革命。効率的なN₂固定 = 肥料のエネルギー使用量の大幅な削減(現在は世界のエネルギー消費の1〜2%)。これらは数兆ドル規模の問題です。",{"type":21,"tag":22,"props":20190,"children":20191},{},[20192,20196],{"type":21,"tag":71,"props":20193,"children":20194},{},[20195],{"type":26,"value":20097},{"type":26,"value":20197}," Microsoft(StationQ)、IBM、そしてQunaSysやGood Chemistryのようなスタートアップがこれを追求しています。初期段階の研究です。",{"type":21,"tag":22,"props":20199,"children":20200},{},[20201,20205],{"type":21,"tag":71,"props":20202,"children":20203},{},[20204],{"type":26,"value":20107},{"type":26,"value":20206}," 目標精度と分子サイズに応じて5〜20年。",{"type":21,"tag":18616,"props":20208,"children":20209},{},[],{"type":21,"tag":59,"props":20211,"children":20213},{"id":20212},"完全なフォールトトレランスを必要とすること720年以上",[20214],{"type":26,"value":20215},"完全なフォールトトレランスを必要とすること(7〜20年以上)",{"type":21,"tag":17800,"props":20217,"children":20219},{"id":20218},"shorのアルゴリズムによるrsaeccの解読",[20220],{"type":26,"value":20221},"ShorのアルゴリズムによるRSA\u002FECCの解読",{"type":21,"tag":22,"props":20223,"children":20224},{},[20225,20229],{"type":21,"tag":71,"props":20226,"children":20227},{},[20228],{"type":26,"value":19671},{"type":26,"value":20230}," RSA-2048の基盤となる大きな素数の因数分解。",{"type":21,"tag":22,"props":20232,"children":20233},{},[20234,20239],{"type":21,"tag":71,"props":20235,"children":20236},{},[20237],{"type":26,"value":20238},"必要なもの:",{"type":26,"value":20240}," 約4,000の論理量子ビット(それぞれに約1,000の物理量子ビットが必要 = 約400万の物理量子ビット)。現状:数百の物理量子ビット、非常にノイズが多い。",{"type":21,"tag":22,"props":20242,"children":20243},{},[20244,20249],{"type":21,"tag":71,"props":20245,"children":20246},{},[20247],{"type":26,"value":20248},"タイムライン:",{"type":26,"value":20250}," 大半の推定では、RSAを脅かす量子コンピューターは2030〜2040年とされています。決して来ないと言う人もいます。この規模での工学的課題は解決不可能かもしれません。",{"type":21,"tag":22,"props":20252,"children":20253},{},[20254,20259,20261,20266],{"type":21,"tag":71,"props":20255,"children":20256},{},[20257],{"type":26,"value":20258},"今すべきこと:",{"type":26,"value":20260}," タイムラインにかかわらず",{"type":21,"tag":29,"props":20262,"children":20263},{"href":9817},[20264],{"type":26,"value":20265},"ポスト量子暗号へ移行する",{"type":26,"value":20267},"こと。移行には数年かかり、HNDL攻撃は現在進行中です。",{"type":21,"tag":18616,"props":20269,"children":20270},{},[],{"type":21,"tag":17800,"props":20272,"children":20274},{"id":20273},"大規模なgroverの探索",[20275],{"type":26,"value":20276},"大規模なGroverの探索",{"type":21,"tag":22,"props":20278,"children":20279},{},[20280,20284],{"type":21,"tag":71,"props":20281,"children":20282},{},[20283],{"type":26,"value":19671},{"type":26,"value":20285}," ソートされていないデータベースの探索、NP困難問題の解決に対する二次的高速化。",{"type":21,"tag":22,"props":20287,"children":20288},{},[20289,20293],{"type":21,"tag":71,"props":20290,"children":20291},{},[20292],{"type":26,"value":20238},{"type":26,"value":20294}," 深い、誤り訂正された回路。二次的高速化とは、大きなNに対して√Nでも依然として膨大であることを意味します。定数因子が量子ハードウェアに有利でなければなりません。",{"type":21,"tag":22,"props":20296,"children":20297},{},[20298,20302],{"type":21,"tag":71,"props":20299,"children":20300},{},[20301],{"type":26,"value":20248},{"type":26,"value":20303}," 長期的な研究。",{"type":21,"tag":18616,"props":20305,"children":20306},{},[],{"type":21,"tag":17800,"props":20308,"children":20310},{"id":20309},"気候モデリングと流体力学",[20311],{"type":26,"value":20309},{"type":21,"tag":22,"props":20313,"children":20314},{},[20315,20319],{"type":21,"tag":71,"props":20316,"children":20317},{},[20318],{"type":26,"value":19671},{"type":26,"value":20320}," 微分方程式を解く量子アルゴリズム(HHLアルゴリズム)は、気候シミュレーション、流体力学、金融モデリングを加速できる可能性があります。",{"type":21,"tag":22,"props":20322,"children":20323},{},[20324,20328],{"type":21,"tag":71,"props":20325,"children":20326},{},[20327],{"type":26,"value":20238},{"type":26,"value":20329}," 完全なフォールトトレランス。HHLアルゴリズムは膨大なオーバーヘッドを持ち、理論上でさえ実用的な優位性が明確とは程遠いものになっています。",{"type":21,"tag":22,"props":20331,"children":20332},{},[20333,20337],{"type":21,"tag":71,"props":20334,"children":20335},{},[20336],{"type":26,"value":20248},{"type":26,"value":20338}," これは最も投機的なカテゴリーです。PDEを解くための量子優位性は、実際にはまだ実証されていません。",{"type":21,"tag":18616,"props":20340,"children":20341},{},[],{"type":21,"tag":59,"props":20343,"children":20345},{"id":20344},"正直な結論",[20346],{"type":26,"value":20344},{"type":21,"tag":4503,"props":20348,"children":20349},{},[20350,20371],{"type":21,"tag":4507,"props":20351,"children":20352},{},[20353],{"type":21,"tag":4511,"props":20354,"children":20355},{},[20356,20361,20366],{"type":21,"tag":4515,"props":20357,"children":20358},{},[20359],{"type":26,"value":20360},"応用",{"type":21,"tag":4515,"props":20362,"children":20363},{},[20364],{"type":26,"value":20365},"今日の量子優位性は?",{"type":21,"tag":4515,"props":20367,"children":20368},{},[20369],{"type":26,"value":20370},"見通し",{"type":21,"tag":4529,"props":20372,"children":20373},{},[20374,20392,20410,20427,20445,20463,20479,20496],{"type":21,"tag":4511,"props":20375,"children":20376},{},[20377,20382,20387],{"type":21,"tag":4536,"props":20378,"children":20379},{},[20380],{"type":26,"value":20381},"ランダム回路サンプリング",{"type":21,"tag":4536,"props":20383,"children":20384},{},[20385],{"type":26,"value":20386},"わずか、応用なし",{"type":21,"tag":4536,"props":20388,"children":20389},{},[20390],{"type":26,"value":20391},"現在(研究のみ)",{"type":21,"tag":4511,"props":20393,"children":20394},{},[20395,20400,20405],{"type":21,"tag":4536,"props":20396,"children":20397},{},[20398],{"type":26,"value":20399},"小分子向けVQE",{"type":21,"tag":4536,"props":20401,"children":20402},{},[20403],{"type":26,"value":20404},"古典に対する優位性なし",{"type":21,"tag":4536,"props":20406,"children":20407},{},[20408],{"type":26,"value":20409},"有用な分子まで3〜5年",{"type":21,"tag":4511,"props":20411,"children":20412},{},[20413,20418,20422],{"type":21,"tag":4536,"props":20414,"children":20415},{},[20416],{"type":26,"value":20417},"最適化向けQAOA",{"type":21,"tag":4536,"props":20419,"children":20420},{},[20421],{"type":26,"value":20404},{"type":21,"tag":4536,"props":20423,"children":20424},{},[20425],{"type":26,"value":20426},"5〜10年",{"type":21,"tag":4511,"props":20428,"children":20429},{},[20430,20435,20440],{"type":21,"tag":4536,"props":20431,"children":20432},{},[20433],{"type":26,"value":20434},"QML \u002F 量子カーネル",{"type":21,"tag":4536,"props":20436,"children":20437},{},[20438],{"type":26,"value":20439},"実データに対する優位性なし",{"type":21,"tag":4536,"props":20441,"children":20442},{},[20443],{"type":26,"value":20444},"不確実",{"type":21,"tag":4511,"props":20446,"children":20447},{},[20448,20453,20458],{"type":21,"tag":4536,"props":20449,"children":20450},{},[20451],{"type":26,"value":20452},"創薬",{"type":21,"tag":4536,"props":20454,"children":20455},{},[20456],{"type":26,"value":20457},"なし",{"type":21,"tag":4536,"props":20459,"children":20460},{},[20461],{"type":26,"value":20462},"7〜10年",{"type":21,"tag":4511,"props":20464,"children":20465},{},[20466,20471,20475],{"type":21,"tag":4536,"props":20467,"children":20468},{},[20469],{"type":26,"value":20470},"金融(モンテカルロ)",{"type":21,"tag":4536,"props":20472,"children":20473},{},[20474],{"type":26,"value":20457},{"type":21,"tag":4536,"props":20476,"children":20477},{},[20478],{"type":26,"value":20426},{"type":21,"tag":4511,"props":20480,"children":20481},{},[20482,20487,20491],{"type":21,"tag":4536,"props":20483,"children":20484},{},[20485],{"type":26,"value":20486},"材料シミュレーション",{"type":21,"tag":4536,"props":20488,"children":20489},{},[20490],{"type":26,"value":20457},{"type":21,"tag":4536,"props":20492,"children":20493},{},[20494],{"type":26,"value":20495},"7〜15年",{"type":21,"tag":4511,"props":20497,"children":20498},{},[20499,20504,20508],{"type":21,"tag":4536,"props":20500,"children":20501},{},[20502],{"type":26,"value":20503},"暗号(Shor)",{"type":21,"tag":4536,"props":20505,"children":20506},{},[20507],{"type":26,"value":20457},{"type":21,"tag":4536,"props":20509,"children":20510},{},[20511],{"type":26,"value":20512},"10〜20年以上",{"type":21,"tag":22,"props":20514,"children":20515},{},[20516,20518,20523,20525,20530],{"type":26,"value":20517},"最も差し迫った量子の価値は、アルゴリズムを実行することに",{"type":21,"tag":71,"props":20519,"children":20520},{},[20521],{"type":26,"value":20522},"ある",{"type":26,"value":20524},"のではなく、",{"type":21,"tag":71,"props":20526,"children":20527},{},[20528],{"type":26,"value":20529},"この分野を早く学ぶこと",{"type":26,"value":20531},"にあります。IBM、Google、NVIDIA、AWS、IonQ、そして数十のスタートアップが量子エンジニアを採用しています。そのスキルセット、すなわち回路設計、変分アルゴリズム、量子古典ハイブリッドワークフローは、本当に希少です。そのギャップはやがて埋まり、早く学んだ人々が有利な立場に立つでしょう。",{"type":21,"tag":22,"props":20533,"children":20534},{},[20535,20537,20542],{"type":26,"value":20536},"2番目に差し迫った価値は",{"type":21,"tag":71,"props":20538,"children":20539},{},[20540],{"type":26,"value":20541},"ポスト量子暗号",{"type":26,"value":20543},"です。これは量子の脅威を伴う古典的なソフトウェアの問題であり、求められる行動は明確かつ緊急です。",{"type":21,"tag":22,"props":20545,"children":20546},{},[20547],{"type":26,"value":20548},"それ以外のすべてについて言えば、ハードウェアは本物であり、進歩は本物であり、誇大宣伝は時期尚早であり、そして軌道は急峻です。",{"type":21,"tag":22,"props":20550,"children":20551},{},[20552,20557,20558,20563,20564,20569,20570],{"type":21,"tag":71,"props":20553,"children":20554},{},[20555],{"type":26,"value":20556},"ハードウェアを探る:",{"type":26,"value":1901},{"type":21,"tag":29,"props":20559,"children":20560},{"href":806},[20561],{"type":26,"value":20562},"無料のQPUアクセス",{"type":26,"value":7549},{"type":21,"tag":29,"props":20565,"children":20566},{"href":1005},[20567],{"type":26,"value":20568},"すべてのシミュレーターを比較",{"type":26,"value":7549},{"type":21,"tag":29,"props":20571,"children":20572},{"href":11943},[20573],{"type":26,"value":20574},"量子ユースケースの詳細",{"type":21,"tag":703,"props":20576,"children":20577},{},[20578],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":20580},[20581,20582,20588,20593,20598],{"id":19603,"depth":335,"text":19603},{"id":19654,"depth":335,"text":19657,"children":20583},[20584,20585,20586,20587],{"id":19660,"depth":344,"text":19663},{"id":19884,"depth":344,"text":19887},{"id":19948,"depth":344,"text":19951},{"id":20009,"depth":344,"text":20012},{"id":20063,"depth":335,"text":20066,"children":20589},[20590,20591,20592],{"id":20069,"depth":344,"text":20069},{"id":20115,"depth":344,"text":20118},{"id":20167,"depth":344,"text":20170},{"id":20212,"depth":335,"text":20215,"children":20594},[20595,20596,20597],{"id":20218,"depth":344,"text":20221},{"id":20273,"depth":344,"text":20276},{"id":20309,"depth":344,"text":20309},{"id":20344,"depth":335,"text":20344},"content:ja:blog:quantum-use-cases-2026.md","ja\u002Fblog\u002Fquantum-use-cases-2026.md","ja\u002Fblog\u002Fquantum-use-cases-2026",{"_path":20603,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":20604,"description":20605,"date":20606,"author":11,"tags":20607,"readingTime":453,"body":20609,"_type":715,"_id":21446,"_source":717,"_file":21447,"_stem":21448,"_extension":720},"\u002Fja\u002Fblog\u002Fgetting-started-free-quantum-computing","2026年に無料で始める量子コンピューティング入門","初めての量子回路を無料で実行するための完全な初心者向けガイド — 高価なハードウェアも、有料サブスクリプションも不要です。","2026-03-10",[7586,20608,1962],"Simulators",{"type":18,"children":20610,"toc":21438},[20611,20616,20621,20626,20661,20666,20671,20676,20701,20713,20923,20958,20964,20969,21043,21063,21069,21081,21106,21192,21200,21205,21211,21325,21336,21341,21346,21395,21413,21416,21434],{"type":21,"tag":22,"props":20612,"children":20613},{},[20614],{"type":26,"value":20615},"量子コンピューティングはもはや数百万ドルの研究所設備の中に閉じ込められたものではありません。今日では、あなたのノートパソコンから、実機の127量子ビットマシン上で量子回路を書いて実行できます — しかも無料で。このガイドでは、今すぐ利用できるあらゆる選択肢を順を追って説明します。",{"type":21,"tag":59,"props":20617,"children":20619},{"id":20618},"実際に必要なもの",[20620],{"type":26,"value":20618},{"type":21,"tag":22,"props":20622,"children":20623},{},[20624],{"type":26,"value":20625},"始めるために必要なのは、ちょうど3つだけです。",{"type":21,"tag":7444,"props":20627,"children":20628},{},[20629,20639,20651],{"type":21,"tag":3560,"props":20630,"children":20631},{},[20632,20634],{"type":26,"value":20633},"マシンにインストールされた ",{"type":21,"tag":71,"props":20635,"children":20636},{},[20637],{"type":26,"value":20638},"Python 3.9+",{"type":21,"tag":3560,"props":20640,"children":20641},{},[20642,20644,20649],{"type":26,"value":20643},"量子クラウドプラットフォームのいずれかの ",{"type":21,"tag":71,"props":20645,"children":20646},{},[20647],{"type":26,"value":20648},"無料アカウント",{"type":26,"value":20650},"（IBM Quantum が最も簡単です）",{"type":21,"tag":3560,"props":20652,"children":20653},{},[20654,20656],{"type":26,"value":20655},"初めての Bell 状態を実行するための ",{"type":21,"tag":71,"props":20657,"children":20658},{},[20659],{"type":26,"value":20660},"5分間",{"type":21,"tag":22,"props":20662,"children":20663},{},[20664],{"type":26,"value":20665},"これだけです。GPU も不要。特別なハードウェアも不要。クレジットカードも不要です。",{"type":21,"tag":59,"props":20667,"children":20669},{"id":20668},"初めての量子回路",[20670],{"type":26,"value":20668},{"type":21,"tag":22,"props":20672,"children":20673},{},[20674],{"type":26,"value":20675},"Qiskit とそのシミュレータをインストールします。",{"type":21,"tag":312,"props":20677,"children":20678},{"className":10181,"code":12220,"language":10183,"meta":7,"style":7},[20679],{"type":21,"tag":79,"props":20680,"children":20681},{"__ignoreMap":7},[20682],{"type":21,"tag":322,"props":20683,"children":20684},{"class":324,"line":325},[20685,20689,20693,20697],{"type":21,"tag":322,"props":20686,"children":20687},{"style":4801},[20688],{"type":26,"value":12232},{"type":21,"tag":322,"props":20690,"children":20691},{"style":359},[20692],{"type":26,"value":12237},{"type":21,"tag":322,"props":20694,"children":20695},{"style":359},[20696],{"type":26,"value":12242},{"type":21,"tag":322,"props":20698,"children":20699},{"style":359},[20700],{"type":26,"value":12247},{"type":21,"tag":22,"props":20702,"children":20703},{},[20704,20706,20711],{"type":26,"value":20705},"では ",{"type":21,"tag":71,"props":20707,"children":20708},{},[20709],{"type":26,"value":20710},"Bell 状態",{"type":26,"value":20712}," を作ってみましょう — 重ね合わせとエンタングルメントを示す、典型的な2量子ビット回路です。",{"type":21,"tag":312,"props":20714,"children":20716},{"className":3626,"code":20715,"language":3628,"meta":7,"style":7},"from qiskit import QuantumCircuit\nfrom qiskit_aer import AerSimulator\n\n# Build the circuit\nqc = QuantumCircuit(2)\nqc.h(0)       # Hadamard — puts qubit 0 into superposition\nqc.cx(0, 1)   # CNOT — entangles qubit 1 with qubit 0\nqc.measure_all()\n\n# Run locally on a simulator\nresult = AerSimulator().run(qc, shots=1000).result()\nprint(result.get_counts())\n# {'00': 503, '11': 497}\n",[20717],{"type":21,"tag":79,"props":20718,"children":20719},{"__ignoreMap":7},[20720,20740,20759,20766,20774,20797,20818,20847,20854,20861,20869,20903,20915],{"type":21,"tag":322,"props":20721,"children":20722},{"class":324,"line":325},[20723,20727,20731,20735],{"type":21,"tag":322,"props":20724,"children":20725},{"style":3638},[20726],{"type":26,"value":3641},{"type":21,"tag":322,"props":20728,"children":20729},{"style":329},[20730],{"type":26,"value":4732},{"type":21,"tag":322,"props":20732,"children":20733},{"style":3638},[20734],{"type":26,"value":3651},{"type":21,"tag":322,"props":20736,"children":20737},{"style":329},[20738],{"type":26,"value":20739}," QuantumCircuit\n",{"type":21,"tag":322,"props":20741,"children":20742},{"class":324,"line":335},[20743,20747,20751,20755],{"type":21,"tag":322,"props":20744,"children":20745},{"style":3638},[20746],{"type":26,"value":3641},{"type":21,"tag":322,"props":20748,"children":20749},{"style":329},[20750],{"type":26,"value":4753},{"type":21,"tag":322,"props":20752,"children":20753},{"style":3638},[20754],{"type":26,"value":3651},{"type":21,"tag":322,"props":20756,"children":20757},{"style":329},[20758],{"type":26,"value":4762},{"type":21,"tag":322,"props":20760,"children":20761},{"class":324,"line":344},[20762],{"type":21,"tag":322,"props":20763,"children":20764},{"emptyLinePlaceholder":3662},[20765],{"type":26,"value":3665},{"type":21,"tag":322,"props":20767,"children":20768},{"class":324,"line":407},[20769],{"type":21,"tag":322,"props":20770,"children":20771},{"style":3888},[20772],{"type":26,"value":20773},"# Build the circuit\n",{"type":21,"tag":322,"props":20775,"children":20776},{"class":324,"line":431},[20777,20781,20785,20789,20793],{"type":21,"tag":322,"props":20778,"children":20779},{"style":329},[20780],{"type":26,"value":7819},{"type":21,"tag":322,"props":20782,"children":20783},{"style":3638},[20784],{"type":26,"value":3678},{"type":21,"tag":322,"props":20786,"children":20787},{"style":329},[20788],{"type":26,"value":5575},{"type":21,"tag":322,"props":20790,"children":20791},{"style":348},[20792],{"type":26,"value":423},{"type":21,"tag":322,"props":20794,"children":20795},{"style":329},[20796],{"type":26,"value":3738},{"type":21,"tag":322,"props":20798,"children":20799},{"class":324,"line":453},[20800,20804,20808,20813],{"type":21,"tag":322,"props":20801,"children":20802},{"style":329},[20803],{"type":26,"value":7835},{"type":21,"tag":322,"props":20805,"children":20806},{"style":348},[20807],{"type":26,"value":5605},{"type":21,"tag":322,"props":20809,"children":20810},{"style":329},[20811],{"type":26,"value":20812},")       ",{"type":21,"tag":322,"props":20814,"children":20815},{"style":3888},[20816],{"type":26,"value":20817},"# Hadamard — puts qubit 0 into superposition\n",{"type":21,"tag":322,"props":20819,"children":20820},{"class":324,"line":474},[20821,20826,20830,20834,20838,20842],{"type":21,"tag":322,"props":20822,"children":20823},{"style":329},[20824],{"type":26,"value":20825},"qc.cx(",{"type":21,"tag":322,"props":20827,"children":20828},{"style":348},[20829],{"type":26,"value":5605},{"type":21,"tag":322,"props":20831,"children":20832},{"style":329},[20833],{"type":26,"value":367},{"type":21,"tag":322,"props":20835,"children":20836},{"style":348},[20837],{"type":26,"value":3994},{"type":21,"tag":322,"props":20839,"children":20840},{"style":329},[20841],{"type":26,"value":7193},{"type":21,"tag":322,"props":20843,"children":20844},{"style":3888},[20845],{"type":26,"value":20846},"# CNOT — entangles qubit 1 with qubit 0\n",{"type":21,"tag":322,"props":20848,"children":20849},{"class":324,"line":496},[20850],{"type":21,"tag":322,"props":20851,"children":20852},{"style":329},[20853],{"type":26,"value":13527},{"type":21,"tag":322,"props":20855,"children":20856},{"class":324,"line":16},[20857],{"type":21,"tag":322,"props":20858,"children":20859},{"emptyLinePlaceholder":3662},[20860],{"type":26,"value":3665},{"type":21,"tag":322,"props":20862,"children":20863},{"class":324,"line":3884},[20864],{"type":21,"tag":322,"props":20865,"children":20866},{"style":3888},[20867],{"type":26,"value":20868},"# Run locally on a simulator\n",{"type":21,"tag":322,"props":20870,"children":20871},{"class":324,"line":3894},[20872,20876,20880,20885,20889,20893,20898],{"type":21,"tag":322,"props":20873,"children":20874},{"style":329},[20875],{"type":26,"value":19764},{"type":21,"tag":322,"props":20877,"children":20878},{"style":3638},[20879],{"type":26,"value":3678},{"type":21,"tag":322,"props":20881,"children":20882},{"style":329},[20883],{"type":26,"value":20884}," AerSimulator().run(qc, ",{"type":21,"tag":322,"props":20886,"children":20887},{"style":3703},[20888],{"type":26,"value":4560},{"type":21,"tag":322,"props":20890,"children":20891},{"style":3638},[20892],{"type":26,"value":3678},{"type":21,"tag":322,"props":20894,"children":20895},{"style":348},[20896],{"type":26,"value":20897},"1000",{"type":21,"tag":322,"props":20899,"children":20900},{"style":329},[20901],{"type":26,"value":20902},").result()\n",{"type":21,"tag":322,"props":20904,"children":20905},{"class":324,"line":1034},[20906,20910],{"type":21,"tag":322,"props":20907,"children":20908},{"style":348},[20909],{"type":26,"value":3753},{"type":21,"tag":322,"props":20911,"children":20912},{"style":329},[20913],{"type":26,"value":20914},"(result.get_counts())\n",{"type":21,"tag":322,"props":20916,"children":20917},{"class":324,"line":3954},[20918],{"type":21,"tag":322,"props":20919,"children":20920},{"style":3888},[20921],{"type":26,"value":20922},"# {'00': 503, '11': 497}\n",{"type":21,"tag":22,"props":20924,"children":20925},{},[20926,20928,20933,20935,20940,20942,20948,20950,20956],{"type":26,"value":20927},"出力 — ",{"type":21,"tag":79,"props":20929,"children":20931},{"className":20930},[],[20932],{"type":26,"value":6113},{"type":26,"value":20934}," と ",{"type":21,"tag":79,"props":20936,"children":20938},{"className":20937},[],[20939],{"type":26,"value":6120},{"type":26,"value":20941}," がほぼ同数で、",{"type":21,"tag":79,"props":20943,"children":20945},{"className":20944},[],[20946],{"type":26,"value":20947},"01",{"type":26,"value":20949}," や ",{"type":21,"tag":79,"props":20951,"children":20953},{"className":20952},[],[20954],{"type":26,"value":20955},"10",{"type":26,"value":20957}," は決して現れない — はエンタングルメントの証です。一方の量子ビットを測定すれば、両者の間の距離に関係なく、もう一方の状態が瞬時に分かります。",{"type":21,"tag":59,"props":20959,"children":20961},{"id":20960},"無料シミュレータ-アカウント不要",[20962],{"type":26,"value":20963},"無料シミュレータ: アカウント不要",{"type":21,"tag":22,"props":20965,"children":20966},{},[20967],{"type":26,"value":20968},"すべての主要な SDK には、サインアップなしで CPU または GPU 上で動作する強力なローカルシミュレータが含まれています。",{"type":21,"tag":312,"props":20970,"children":20972},{"className":10181,"code":20971,"language":10183,"meta":7,"style":7},"pip install qiskit qiskit-aer   # IBM Qiskit + Aer simulator\npip install cirq                 # Google Cirq\npip install pennylane            # PennyLane\n",[20973],{"type":21,"tag":79,"props":20974,"children":20975},{"__ignoreMap":7},[20976,21001,21022],{"type":21,"tag":322,"props":20977,"children":20978},{"class":324,"line":325},[20979,20983,20987,20991,20996],{"type":21,"tag":322,"props":20980,"children":20981},{"style":4801},[20982],{"type":26,"value":12232},{"type":21,"tag":322,"props":20984,"children":20985},{"style":359},[20986],{"type":26,"value":12237},{"type":21,"tag":322,"props":20988,"children":20989},{"style":359},[20990],{"type":26,"value":12242},{"type":21,"tag":322,"props":20992,"children":20993},{"style":359},[20994],{"type":26,"value":20995}," qiskit-aer",{"type":21,"tag":322,"props":20997,"children":20998},{"style":3888},[20999],{"type":26,"value":21000},"   # IBM Qiskit + Aer simulator\n",{"type":21,"tag":322,"props":21002,"children":21003},{"class":324,"line":335},[21004,21008,21012,21017],{"type":21,"tag":322,"props":21005,"children":21006},{"style":4801},[21007],{"type":26,"value":12232},{"type":21,"tag":322,"props":21009,"children":21010},{"style":359},[21011],{"type":26,"value":12237},{"type":21,"tag":322,"props":21013,"children":21014},{"style":359},[21015],{"type":26,"value":21016}," cirq",{"type":21,"tag":322,"props":21018,"children":21019},{"style":3888},[21020],{"type":26,"value":21021},"                 # Google Cirq\n",{"type":21,"tag":322,"props":21023,"children":21024},{"class":324,"line":344},[21025,21029,21033,21038],{"type":21,"tag":322,"props":21026,"children":21027},{"style":4801},[21028],{"type":26,"value":12232},{"type":21,"tag":322,"props":21030,"children":21031},{"style":359},[21032],{"type":26,"value":12237},{"type":21,"tag":322,"props":21034,"children":21035},{"style":359},[21036],{"type":26,"value":21037}," pennylane",{"type":21,"tag":322,"props":21039,"children":21040},{"style":3888},[21041],{"type":26,"value":21042},"            # PennyLane\n",{"type":21,"tag":22,"props":21044,"children":21045},{},[21046,21048,21054,21056,21061],{"type":26,"value":21047},"いずれも20〜30量子ビットの回路を余裕をもって扱えます。大規模な回路や GPU アクセラレーションには、",{"type":21,"tag":29,"props":21049,"children":21051},{"href":21050},"\u002Fsdks\u002Fcudaq",[21052],{"type":26,"value":21053},"NVIDIA CUDA-Q",{"type":26,"value":21055}," が単一の GPU 上で34量子ビット以上をシミュレートできます。詳しい比較については、",{"type":21,"tag":29,"props":21057,"children":21058},{"href":1005},[21059],{"type":26,"value":21060},"Simulators ガイド",{"type":26,"value":21062},"の全文をご覧ください。",{"type":21,"tag":59,"props":21064,"children":21066},{"id":21065},"無料の実機ハードウェア-ibm-quantum",[21067],{"type":26,"value":21068},"無料の実機ハードウェア: IBM Quantum",{"type":21,"tag":22,"props":21070,"children":21071},{},[21072,21074,21079],{"type":26,"value":21073},"IBM Quantum は、IBM ID を持つ誰に対しても、実機の量子プロセッサへの ",{"type":21,"tag":71,"props":21075,"children":21076},{},[21077],{"type":26,"value":21078},"完全無料",{"type":26,"value":21080}," アクセスを提供しています。",{"type":21,"tag":7444,"props":21082,"children":21083},{},[21084,21096,21101],{"type":21,"tag":3560,"props":21085,"children":21086},{},[21087,21094],{"type":21,"tag":29,"props":21088,"children":21091},{"href":21089,"rel":21090},"https:\u002F\u002Fquantum.ibm.com\u002F",[44],[21092],{"type":26,"value":21093},"quantum.ibm.com",{"type":26,"value":21095}," にアクセスし、無料アカウントを作成します",{"type":21,"tag":3560,"props":21097,"children":21098},{},[21099],{"type":26,"value":21100},"アカウント設定から API トークンをコピーします",{"type":21,"tag":3560,"props":21102,"children":21103},{},[21104],{"type":26,"value":21105},"トークンで Qiskit を設定します。",{"type":21,"tag":312,"props":21107,"children":21109},{"className":3626,"code":21108,"language":3628,"meta":7,"style":7},"from qiskit_ibm_runtime import QiskitRuntimeService\n\nQiskitRuntimeService.save_account(\n    channel=\"ibm_quantum\",\n    token=\"YOUR_TOKEN_HERE\"\n)\n",[21110],{"type":21,"tag":79,"props":21111,"children":21112},{"__ignoreMap":7},[21113,21132,21139,21147,21168,21185],{"type":21,"tag":322,"props":21114,"children":21115},{"class":324,"line":325},[21116,21120,21124,21128],{"type":21,"tag":322,"props":21117,"children":21118},{"style":3638},[21119],{"type":26,"value":3641},{"type":21,"tag":322,"props":21121,"children":21122},{"style":329},[21123],{"type":26,"value":3646},{"type":21,"tag":322,"props":21125,"children":21126},{"style":3638},[21127],{"type":26,"value":3651},{"type":21,"tag":322,"props":21129,"children":21130},{"style":329},[21131],{"type":26,"value":3656},{"type":21,"tag":322,"props":21133,"children":21134},{"class":324,"line":335},[21135],{"type":21,"tag":322,"props":21136,"children":21137},{"emptyLinePlaceholder":3662},[21138],{"type":26,"value":3665},{"type":21,"tag":322,"props":21140,"children":21141},{"class":324,"line":344},[21142],{"type":21,"tag":322,"props":21143,"children":21144},{"style":329},[21145],{"type":26,"value":21146},"QiskitRuntimeService.save_account(\n",{"type":21,"tag":322,"props":21148,"children":21149},{"class":324,"line":407},[21150,21155,21159,21164],{"type":21,"tag":322,"props":21151,"children":21152},{"style":3703},[21153],{"type":26,"value":21154},"    channel",{"type":21,"tag":322,"props":21156,"children":21157},{"style":3638},[21158],{"type":26,"value":3678},{"type":21,"tag":322,"props":21160,"children":21161},{"style":359},[21162],{"type":26,"value":21163},"\"ibm_quantum\"",{"type":21,"tag":322,"props":21165,"children":21166},{"style":329},[21167],{"type":26,"value":428},{"type":21,"tag":322,"props":21169,"children":21170},{"class":324,"line":431},[21171,21176,21180],{"type":21,"tag":322,"props":21172,"children":21173},{"style":3703},[21174],{"type":26,"value":21175},"    token",{"type":21,"tag":322,"props":21177,"children":21178},{"style":3638},[21179],{"type":26,"value":3678},{"type":21,"tag":322,"props":21181,"children":21182},{"style":359},[21183],{"type":26,"value":21184},"\"YOUR_TOKEN_HERE\"\n",{"type":21,"tag":322,"props":21186,"children":21187},{"class":324,"line":453},[21188],{"type":21,"tag":322,"props":21189,"children":21190},{"style":329},[21191],{"type":26,"value":3738},{"type":21,"tag":7444,"props":21193,"children":21194},{"start":407},[21195],{"type":21,"tag":3560,"props":21196,"children":21197},{},[21198],{"type":26,"value":21199},"IBM の無料公開 QPU のいずれかにジョブを送信します",{"type":21,"tag":22,"props":21201,"children":21202},{},[21203],{"type":26,"value":21204},"公開システムでの待ち時間は通常、数分から数時間です。世界中の研究者や学生と時間を共有することになります — それが唯一の難点です。",{"type":21,"tag":59,"props":21206,"children":21208},{"id":21207},"最初の-sdk-を選ぶ",[21209],{"type":26,"value":21210},"最初の SDK を選ぶ",{"type":21,"tag":4503,"props":21212,"children":21213},{},[21214,21235],{"type":21,"tag":4507,"props":21215,"children":21216},{},[21217],{"type":21,"tag":4511,"props":21218,"children":21219},{},[21220,21225,21230],{"type":21,"tag":4515,"props":21221,"children":21222},{},[21223],{"type":26,"value":21224},"SDK",{"type":21,"tag":4515,"props":21226,"children":21227},{},[21228],{"type":26,"value":21229},"最適な用途",{"type":21,"tag":4515,"props":21231,"children":21232},{},[21233],{"type":26,"value":21234},"無料枠",{"type":21,"tag":4529,"props":21236,"children":21237},{},[21238,21256,21273,21290,21308],{"type":21,"tag":4511,"props":21239,"children":21240},{},[21241,21246,21251],{"type":21,"tag":4536,"props":21242,"children":21243},{},[21244],{"type":26,"value":21245},"IBM Qiskit",{"type":21,"tag":4536,"props":21247,"children":21248},{},[21249],{"type":26,"value":21250},"汎用、実機ハードウェア",{"type":21,"tag":4536,"props":21252,"children":21253},{},[21254],{"type":26,"value":21255},"フル — ローカル + 実機 QPU",{"type":21,"tag":4511,"props":21257,"children":21258},{},[21259,21263,21268],{"type":21,"tag":4536,"props":21260,"children":21261},{},[21262],{"type":26,"value":14223},{"type":21,"tag":4536,"props":21264,"children":21265},{},[21266],{"type":26,"value":21267},"量子機械学習、勾配計算",{"type":21,"tag":4536,"props":21269,"children":21270},{},[21271],{"type":26,"value":21272},"フル — ローカルシミュレータ",{"type":21,"tag":4511,"props":21274,"children":21275},{},[21276,21281,21286],{"type":21,"tag":4536,"props":21277,"children":21278},{},[21279],{"type":26,"value":21280},"Google Cirq",{"type":21,"tag":4536,"props":21282,"children":21283},{},[21284],{"type":26,"value":21285},"NISQ 研究",{"type":21,"tag":4536,"props":21287,"children":21288},{},[21289],{"type":26,"value":21272},{"type":21,"tag":4511,"props":21291,"children":21292},{},[21293,21298,21303],{"type":21,"tag":4536,"props":21294,"children":21295},{},[21296],{"type":26,"value":21297},"Amazon Braket",{"type":21,"tag":4536,"props":21299,"children":21300},{},[21301],{"type":26,"value":21302},"AWS 連携",{"type":21,"tag":4536,"props":21304,"children":21305},{},[21306],{"type":26,"value":21307},"無料ローカルシミュレータ、クラウドは有料",{"type":21,"tag":4511,"props":21309,"children":21310},{},[21311,21315,21320],{"type":21,"tag":4536,"props":21312,"children":21313},{},[21314],{"type":26,"value":21053},{"type":21,"tag":4536,"props":21316,"children":21317},{},[21318],{"type":26,"value":21319},"GPU アクセラレーション",{"type":21,"tag":4536,"props":21321,"children":21322},{},[21323],{"type":26,"value":21324},"フル — ローカル GPU",{"type":21,"tag":22,"props":21326,"children":21327},{},[21328,21330,21334],{"type":26,"value":21329},"ほとんどの初心者にとって、",{"type":21,"tag":71,"props":21331,"children":21332},{},[21333],{"type":26,"value":1962},{"type":26,"value":21335}," が最良の出発点です — 最大のコミュニティ、最も充実したドキュメント、そして IBM 実機ハードウェアへの直接アクセスを備えています。",{"type":21,"tag":59,"props":21337,"children":21339},{"id":21338},"次に作るべきもの",[21340],{"type":26,"value":21338},{"type":21,"tag":22,"props":21342,"children":21343},{},[21344],{"type":26,"value":21345},"Bell 状態が動いたら、次に試すべき興味深い回路の進め方を紹介します。",{"type":21,"tag":3556,"props":21347,"children":21348},{},[21349,21367,21376,21386],{"type":21,"tag":3560,"props":21350,"children":21351},{},[21352,21357,21359,21365],{"type":21,"tag":71,"props":21353,"children":21354},{},[21355],{"type":26,"value":21356},"GHZ 状態",{"type":26,"value":21358}," — 3量子ビットのエンタングル状態（",{"type":21,"tag":79,"props":21360,"children":21362},{"className":21361},[],[21363],{"type":26,"value":21364},"h(0).cx(0,1).cx(0,2)",{"type":26,"value":21366},"）",{"type":21,"tag":3560,"props":21368,"children":21369},{},[21370,21374],{"type":21,"tag":71,"props":21371,"children":21372},{},[21373],{"type":26,"value":4281},{"type":26,"value":21375}," — エンタングルメントと古典通信を用いて量子ビットの状態を転送する",{"type":21,"tag":3560,"props":21377,"children":21378},{},[21379,21384],{"type":21,"tag":71,"props":21380,"children":21381},{},[21382],{"type":26,"value":21383},"Grover の探索",{"type":26,"value":21385}," — リスト内のマークされた項目を、総当たりよりも二次的に速く見つける",{"type":21,"tag":3560,"props":21387,"children":21388},{},[21389,21393],{"type":21,"tag":71,"props":21390,"children":21391},{},[21392],{"type":26,"value":883},{"type":26,"value":21394}," — 変分アルゴリズムで分子の基底状態エネルギーを推定する",{"type":21,"tag":22,"props":21396,"children":21397},{},[21398,21400,21404,21406,21411],{"type":26,"value":21399},"さらに深く学ぶには、",{"type":21,"tag":29,"props":21401,"children":21402},{"href":1005},[21403],{"type":26,"value":21060},{"type":26,"value":21405},"と ",{"type":21,"tag":29,"props":21407,"children":21408},{"href":806},[21409],{"type":26,"value":21410},"Hardware ガイド",{"type":26,"value":21412},"をご覧ください。",{"type":21,"tag":18616,"props":21414,"children":21415},{},[],{"type":21,"tag":22,"props":21417,"children":21418},{},[21419,21421,21426,21428,21432],{"type":26,"value":21420},"💡 ",{"type":21,"tag":71,"props":21422,"children":21423},{},[21424],{"type":26,"value":21425},"ヒント:",{"type":26,"value":21427}," 複数の SDK にまたがって、毎回回路を書き直すことなく実験したい場合は、",{"type":21,"tag":29,"props":21429,"children":21430},{"href":19441},[21431],{"type":26,"value":19444},{"type":26,"value":21433}," が統一 API を提供しています。単一のバックエンドフラグで、同じ回路を Qiskit、Cirq、PennyLane、Braket、CUDA-Q、または IonQ 上で実行できます。",{"type":21,"tag":703,"props":21435,"children":21436},{},[21437],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":21439},[21440,21441,21442,21443,21444,21445],{"id":20618,"depth":335,"text":20618},{"id":20668,"depth":335,"text":20668},{"id":20960,"depth":335,"text":20963},{"id":21065,"depth":335,"text":21068},{"id":21207,"depth":335,"text":21210},{"id":21338,"depth":335,"text":21338},"content:ja:blog:getting-started-free-quantum-computing.md","ja\u002Fblog\u002Fgetting-started-free-quantum-computing.md","ja\u002Fblog\u002Fgetting-started-free-quantum-computing",{"_path":21450,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":21451,"description":21452,"date":21453,"author":11,"tags":21454,"readingTime":474,"body":21456,"_type":715,"_id":23912,"_source":717,"_file":23913,"_stem":23914,"_extension":720},"\u002Fja\u002Fblog\u002Fhlquantum-vs-qiskit","HLQuantum vs 素のQiskit: 抽象化レイヤーをいつ使うべきか","Qiskitで量子回路を直接記述する場合と、HLQuantumの統一APIを使う場合の実践的な比較 — 並列コード例と、どちらを使うか判断するためのフレームワークとともに。","2026-03-08",[19444,1962,7586,21455],"Best Practices",{"type":18,"children":21457,"toc":23902},[21458,21463,21468,21473,21482,21491,21497,21505,21737,21745,21906,21911,21917,21925,22441,22449,22786,22791,22797,22805,23196,23204,23447,23452,23458,23463,23545,23551,23556,23630,23636,23641,23850,23855,23860,23868,23873,23898],{"type":21,"tag":22,"props":21459,"children":21460},{},[21461],{"type":26,"value":21462},"量子コンピューティングのプロジェクトを始めるとき、避けて通れない問いがあります。回路をQiskitで直接記述すべきか、それともその上にあるHLQuantumの抽象化レイヤーを使うべきか。どちらも妥当な選択であり、正解はあなたが何を作ろうとしているかによって変わります。",{"type":21,"tag":22,"props":21464,"children":21465},{},[21466],{"type":26,"value":21467},"この記事では、判断の助けとなるよう、具体的でコード中心の比較を提供します。",{"type":21,"tag":59,"props":21469,"children":21471},{"id":21470},"核心的な違い",[21472],{"type":26,"value":21470},{"type":21,"tag":22,"props":21474,"children":21475},{},[21476,21480],{"type":21,"tag":71,"props":21477,"children":21478},{},[21479],{"type":26,"value":1962},{"type":26,"value":21481}," はIBMの量子SDKです。すべてのゲート、量子ビット、測定、コンパイルステップに対して直接的かつ精密な制御を提供します。IBMハードウェアをネイティブにターゲットとし、IBM固有のワークフローに対して最も深いエコシステムを備えています。",{"type":21,"tag":22,"props":21483,"children":21484},{},[21485,21489],{"type":21,"tag":71,"props":21486,"children":21487},{},[21488],{"type":26,"value":19444},{"type":26,"value":21490}," は、Qiskit、Cirq、PennyLane、Braket、CUDA-Q、IonQの上に位置する統一APIです。回路を一度記述すれば、指定したバックエンドへ変換してディスパッチします。低レベルの制御を一部犠牲にする代わりに、移植性とシンプルさを得られます。",{"type":21,"tag":59,"props":21492,"children":21494},{"id":21493},"並列比較-ベル状態",[21495],{"type":26,"value":21496},"並列比較: ベル状態",{"type":21,"tag":22,"props":21498,"children":21499},{},[21500],{"type":21,"tag":71,"props":21501,"children":21502},{},[21503],{"type":26,"value":21504},"素のQiskit:",{"type":21,"tag":312,"props":21506,"children":21508},{"code":21507,"language":3628,"meta":7,"className":3626,"style":7},"from qiskit import QuantumCircuit\nfrom qiskit_aer import AerSimulator\n\nqc = QuantumCircuit(2, 2)\nqc.h(0)\nqc.cx(0, 1)\nqc.measure([0, 1], [0, 1])\n\nsim = AerSimulator()\nresult = sim.run(qc, shots=1000).result()\nprint(result.get_counts())  # {'00': 501, '11': 499}\n",[21509],{"type":21,"tag":79,"props":21510,"children":21511},{"__ignoreMap":7},[21512,21531,21550,21557,21588,21603,21626,21666,21673,21688,21720],{"type":21,"tag":322,"props":21513,"children":21514},{"class":324,"line":325},[21515,21519,21523,21527],{"type":21,"tag":322,"props":21516,"children":21517},{"style":3638},[21518],{"type":26,"value":3641},{"type":21,"tag":322,"props":21520,"children":21521},{"style":329},[21522],{"type":26,"value":4732},{"type":21,"tag":322,"props":21524,"children":21525},{"style":3638},[21526],{"type":26,"value":3651},{"type":21,"tag":322,"props":21528,"children":21529},{"style":329},[21530],{"type":26,"value":20739},{"type":21,"tag":322,"props":21532,"children":21533},{"class":324,"line":335},[21534,21538,21542,21546],{"type":21,"tag":322,"props":21535,"children":21536},{"style":3638},[21537],{"type":26,"value":3641},{"type":21,"tag":322,"props":21539,"children":21540},{"style":329},[21541],{"type":26,"value":4753},{"type":21,"tag":322,"props":21543,"children":21544},{"style":3638},[21545],{"type":26,"value":3651},{"type":21,"tag":322,"props":21547,"children":21548},{"style":329},[21549],{"type":26,"value":4762},{"type":21,"tag":322,"props":21551,"children":21552},{"class":324,"line":344},[21553],{"type":21,"tag":322,"props":21554,"children":21555},{"emptyLinePlaceholder":3662},[21556],{"type":26,"value":3665},{"type":21,"tag":322,"props":21558,"children":21559},{"class":324,"line":407},[21560,21564,21568,21572,21576,21580,21584],{"type":21,"tag":322,"props":21561,"children":21562},{"style":329},[21563],{"type":26,"value":7819},{"type":21,"tag":322,"props":21565,"children":21566},{"style":3638},[21567],{"type":26,"value":3678},{"type":21,"tag":322,"props":21569,"children":21570},{"style":329},[21571],{"type":26,"value":5575},{"type":21,"tag":322,"props":21573,"children":21574},{"style":348},[21575],{"type":26,"value":423},{"type":21,"tag":322,"props":21577,"children":21578},{"style":329},[21579],{"type":26,"value":367},{"type":21,"tag":322,"props":21581,"children":21582},{"style":348},[21583],{"type":26,"value":423},{"type":21,"tag":322,"props":21585,"children":21586},{"style":329},[21587],{"type":26,"value":3738},{"type":21,"tag":322,"props":21589,"children":21590},{"class":324,"line":431},[21591,21595,21599],{"type":21,"tag":322,"props":21592,"children":21593},{"style":329},[21594],{"type":26,"value":7835},{"type":21,"tag":322,"props":21596,"children":21597},{"style":348},[21598],{"type":26,"value":5605},{"type":21,"tag":322,"props":21600,"children":21601},{"style":329},[21602],{"type":26,"value":3738},{"type":21,"tag":322,"props":21604,"children":21605},{"class":324,"line":453},[21606,21610,21614,21618,21622],{"type":21,"tag":322,"props":21607,"children":21608},{"style":329},[21609],{"type":26,"value":20825},{"type":21,"tag":322,"props":21611,"children":21612},{"style":348},[21613],{"type":26,"value":5605},{"type":21,"tag":322,"props":21615,"children":21616},{"style":329},[21617],{"type":26,"value":367},{"type":21,"tag":322,"props":21619,"children":21620},{"style":348},[21621],{"type":26,"value":3994},{"type":21,"tag":322,"props":21623,"children":21624},{"style":329},[21625],{"type":26,"value":3738},{"type":21,"tag":322,"props":21627,"children":21628},{"class":324,"line":474},[21629,21634,21638,21642,21646,21650,21654,21658,21662],{"type":21,"tag":322,"props":21630,"children":21631},{"style":329},[21632],{"type":26,"value":21633},"qc.measure([",{"type":21,"tag":322,"props":21635,"children":21636},{"style":348},[21637],{"type":26,"value":5605},{"type":21,"tag":322,"props":21639,"children":21640},{"style":329},[21641],{"type":26,"value":367},{"type":21,"tag":322,"props":21643,"children":21644},{"style":348},[21645],{"type":26,"value":3994},{"type":21,"tag":322,"props":21647,"children":21648},{"style":329},[21649],{"type":26,"value":5660},{"type":21,"tag":322,"props":21651,"children":21652},{"style":348},[21653],{"type":26,"value":5605},{"type":21,"tag":322,"props":21655,"children":21656},{"style":329},[21657],{"type":26,"value":367},{"type":21,"tag":322,"props":21659,"children":21660},{"style":348},[21661],{"type":26,"value":3994},{"type":21,"tag":322,"props":21663,"children":21664},{"style":329},[21665],{"type":26,"value":3999},{"type":21,"tag":322,"props":21667,"children":21668},{"class":324,"line":496},[21669],{"type":21,"tag":322,"props":21670,"children":21671},{"emptyLinePlaceholder":3662},[21672],{"type":26,"value":3665},{"type":21,"tag":322,"props":21674,"children":21675},{"class":324,"line":16},[21676,21680,21684],{"type":21,"tag":322,"props":21677,"children":21678},{"style":329},[21679],{"type":26,"value":8861},{"type":21,"tag":322,"props":21681,"children":21682},{"style":3638},[21683],{"type":26,"value":3678},{"type":21,"tag":322,"props":21685,"children":21686},{"style":329},[21687],{"type":26,"value":8870},{"type":21,"tag":322,"props":21689,"children":21690},{"class":324,"line":3884},[21691,21695,21699,21704,21708,21712,21716],{"type":21,"tag":322,"props":21692,"children":21693},{"style":329},[21694],{"type":26,"value":19764},{"type":21,"tag":322,"props":21696,"children":21697},{"style":3638},[21698],{"type":26,"value":3678},{"type":21,"tag":322,"props":21700,"children":21701},{"style":329},[21702],{"type":26,"value":21703}," sim.run(qc, ",{"type":21,"tag":322,"props":21705,"children":21706},{"style":3703},[21707],{"type":26,"value":4560},{"type":21,"tag":322,"props":21709,"children":21710},{"style":3638},[21711],{"type":26,"value":3678},{"type":21,"tag":322,"props":21713,"children":21714},{"style":348},[21715],{"type":26,"value":20897},{"type":21,"tag":322,"props":21717,"children":21718},{"style":329},[21719],{"type":26,"value":20902},{"type":21,"tag":322,"props":21721,"children":21722},{"class":324,"line":3894},[21723,21727,21732],{"type":21,"tag":322,"props":21724,"children":21725},{"style":348},[21726],{"type":26,"value":3753},{"type":21,"tag":322,"props":21728,"children":21729},{"style":329},[21730],{"type":26,"value":21731},"(result.get_counts())  ",{"type":21,"tag":322,"props":21733,"children":21734},{"style":3888},[21735],{"type":26,"value":21736},"# {'00': 501, '11': 499}\n",{"type":21,"tag":22,"props":21738,"children":21739},{},[21740],{"type":21,"tag":71,"props":21741,"children":21742},{},[21743],{"type":26,"value":21744},"HLQuantum:",{"type":21,"tag":312,"props":21746,"children":21748},{"code":21747,"language":3628,"meta":7,"className":3626,"style":7},"import hlquantum as hlq\n\nqc = hlq.Circuit(2)\nqc.h(0).cx(0, 1).measure_all()\n\nresult = hlq.run(qc, backend='qiskit', shots=1000)\nprint(result.counts)  # {'00': 501, '11': 499}\n",[21749],{"type":21,"tag":79,"props":21750,"children":21751},{"__ignoreMap":7},[21752,21771,21778,21802,21835,21842,21890],{"type":21,"tag":322,"props":21753,"children":21754},{"class":324,"line":325},[21755,21759,21763,21767],{"type":21,"tag":322,"props":21756,"children":21757},{"style":3638},[21758],{"type":26,"value":3651},{"type":21,"tag":322,"props":21760,"children":21761},{"style":329},[21762],{"type":26,"value":19732},{"type":21,"tag":322,"props":21764,"children":21765},{"style":3638},[21766],{"type":26,"value":4715},{"type":21,"tag":322,"props":21768,"children":21769},{"style":329},[21770],{"type":26,"value":19741},{"type":21,"tag":322,"props":21772,"children":21773},{"class":324,"line":335},[21774],{"type":21,"tag":322,"props":21775,"children":21776},{"emptyLinePlaceholder":3662},[21777],{"type":26,"value":3665},{"type":21,"tag":322,"props":21779,"children":21780},{"class":324,"line":344},[21781,21785,21789,21794,21798],{"type":21,"tag":322,"props":21782,"children":21783},{"style":329},[21784],{"type":26,"value":7819},{"type":21,"tag":322,"props":21786,"children":21787},{"style":3638},[21788],{"type":26,"value":3678},{"type":21,"tag":322,"props":21790,"children":21791},{"style":329},[21792],{"type":26,"value":21793}," hlq.Circuit(",{"type":21,"tag":322,"props":21795,"children":21796},{"style":348},[21797],{"type":26,"value":423},{"type":21,"tag":322,"props":21799,"children":21800},{"style":329},[21801],{"type":26,"value":3738},{"type":21,"tag":322,"props":21803,"children":21804},{"class":324,"line":407},[21805,21809,21813,21818,21822,21826,21830],{"type":21,"tag":322,"props":21806,"children":21807},{"style":329},[21808],{"type":26,"value":7835},{"type":21,"tag":322,"props":21810,"children":21811},{"style":348},[21812],{"type":26,"value":5605},{"type":21,"tag":322,"props":21814,"children":21815},{"style":329},[21816],{"type":26,"value":21817},").cx(",{"type":21,"tag":322,"props":21819,"children":21820},{"style":348},[21821],{"type":26,"value":5605},{"type":21,"tag":322,"props":21823,"children":21824},{"style":329},[21825],{"type":26,"value":367},{"type":21,"tag":322,"props":21827,"children":21828},{"style":348},[21829],{"type":26,"value":3994},{"type":21,"tag":322,"props":21831,"children":21832},{"style":329},[21833],{"type":26,"value":21834},").measure_all()\n",{"type":21,"tag":322,"props":21836,"children":21837},{"class":324,"line":431},[21838],{"type":21,"tag":322,"props":21839,"children":21840},{"emptyLinePlaceholder":3662},[21841],{"type":26,"value":3665},{"type":21,"tag":322,"props":21843,"children":21844},{"class":324,"line":453},[21845,21849,21853,21858,21862,21866,21870,21874,21878,21882,21886],{"type":21,"tag":322,"props":21846,"children":21847},{"style":329},[21848],{"type":26,"value":19764},{"type":21,"tag":322,"props":21850,"children":21851},{"style":3638},[21852],{"type":26,"value":3678},{"type":21,"tag":322,"props":21854,"children":21855},{"style":329},[21856],{"type":26,"value":21857}," hlq.run(qc, ",{"type":21,"tag":322,"props":21859,"children":21860},{"style":3703},[21861],{"type":26,"value":18908},{"type":21,"tag":322,"props":21863,"children":21864},{"style":3638},[21865],{"type":26,"value":3678},{"type":21,"tag":322,"props":21867,"children":21868},{"style":359},[21869],{"type":26,"value":19822},{"type":21,"tag":322,"props":21871,"children":21872},{"style":329},[21873],{"type":26,"value":367},{"type":21,"tag":322,"props":21875,"children":21876},{"style":3703},[21877],{"type":26,"value":4560},{"type":21,"tag":322,"props":21879,"children":21880},{"style":3638},[21881],{"type":26,"value":3678},{"type":21,"tag":322,"props":21883,"children":21884},{"style":348},[21885],{"type":26,"value":20897},{"type":21,"tag":322,"props":21887,"children":21888},{"style":329},[21889],{"type":26,"value":3738},{"type":21,"tag":322,"props":21891,"children":21892},{"class":324,"line":474},[21893,21897,21902],{"type":21,"tag":322,"props":21894,"children":21895},{"style":348},[21896],{"type":26,"value":3753},{"type":21,"tag":322,"props":21898,"children":21899},{"style":329},[21900],{"type":26,"value":21901},"(result.counts)  ",{"type":21,"tag":322,"props":21903,"children":21904},{"style":3888},[21905],{"type":26,"value":21736},{"type":21,"tag":22,"props":21907,"children":21908},{},[21909],{"type":26,"value":21910},"単一バックエンドであれば、記述量はほぼ同じです。違いが現れるのは、バックエンドを増やしたときです。",{"type":21,"tag":59,"props":21912,"children":21914},{"id":21913},"並列比較-複数バックエンドでの実行",[21915],{"type":26,"value":21916},"並列比較: 複数バックエンドでの実行",{"type":21,"tag":22,"props":21918,"children":21919},{},[21920],{"type":21,"tag":71,"props":21921,"children":21922},{},[21923],{"type":26,"value":21924},"素のQiskit(2つ目のバックエンドを追加するには、書き直しが必要):",{"type":21,"tag":312,"props":21926,"children":21928},{"code":21927,"language":3628,"meta":7,"className":3626,"style":7},"# Qiskit Aer\nfrom qiskit import QuantumCircuit\nfrom qiskit_aer import AerSimulator\nqc_qiskit = QuantumCircuit(2, 2)\nqc_qiskit.h(0); qc_qiskit.cx(0, 1); qc_qiskit.measure_all()\nresult_qiskit = AerSimulator().run(qc_qiskit, shots=1000).result()\n\n# PennyLane — 全く異なるAPI、回路を書き直す\nimport pennylane as qml\ndev = qml.device('default.qubit', wires=2)\n@qml.qnode(dev)\ndef bell():\n    qml.Hadamard(wires=0)\n    qml.CNOT(wires=[0, 1])\n    return qml.probs(wires=[0, 1])\nresult_pennylane = bell()\n\n# CUDA-Q — さらに別のAPI、また書き直す\nimport cudaq\n@cudaq.kernel\ndef bell():\n    qvec = cudaq.qvector(2)\n    h(qvec[0]); cx(qvec[0], qvec[1]); mz(qvec)\nresult_cudaq = cudaq.sample(bell, shots_count=1000)\n",[21929],{"type":21,"tag":79,"props":21930,"children":21931},{"__ignoreMap":7},[21932,21940,21959,21978,22010,22044,22077,22084,22092,22111,22151,22162,22179,22203,22240,22280,22297,22304,22312,22324,22332,22347,22372,22407],{"type":21,"tag":322,"props":21933,"children":21934},{"class":324,"line":325},[21935],{"type":21,"tag":322,"props":21936,"children":21937},{"style":3888},[21938],{"type":26,"value":21939},"# Qiskit Aer\n",{"type":21,"tag":322,"props":21941,"children":21942},{"class":324,"line":335},[21943,21947,21951,21955],{"type":21,"tag":322,"props":21944,"children":21945},{"style":3638},[21946],{"type":26,"value":3641},{"type":21,"tag":322,"props":21948,"children":21949},{"style":329},[21950],{"type":26,"value":4732},{"type":21,"tag":322,"props":21952,"children":21953},{"style":3638},[21954],{"type":26,"value":3651},{"type":21,"tag":322,"props":21956,"children":21957},{"style":329},[21958],{"type":26,"value":20739},{"type":21,"tag":322,"props":21960,"children":21961},{"class":324,"line":344},[21962,21966,21970,21974],{"type":21,"tag":322,"props":21963,"children":21964},{"style":3638},[21965],{"type":26,"value":3641},{"type":21,"tag":322,"props":21967,"children":21968},{"style":329},[21969],{"type":26,"value":4753},{"type":21,"tag":322,"props":21971,"children":21972},{"style":3638},[21973],{"type":26,"value":3651},{"type":21,"tag":322,"props":21975,"children":21976},{"style":329},[21977],{"type":26,"value":4762},{"type":21,"tag":322,"props":21979,"children":21980},{"class":324,"line":407},[21981,21986,21990,21994,21998,22002,22006],{"type":21,"tag":322,"props":21982,"children":21983},{"style":329},[21984],{"type":26,"value":21985},"qc_qiskit ",{"type":21,"tag":322,"props":21987,"children":21988},{"style":3638},[21989],{"type":26,"value":3678},{"type":21,"tag":322,"props":21991,"children":21992},{"style":329},[21993],{"type":26,"value":5575},{"type":21,"tag":322,"props":21995,"children":21996},{"style":348},[21997],{"type":26,"value":423},{"type":21,"tag":322,"props":21999,"children":22000},{"style":329},[22001],{"type":26,"value":367},{"type":21,"tag":322,"props":22003,"children":22004},{"style":348},[22005],{"type":26,"value":423},{"type":21,"tag":322,"props":22007,"children":22008},{"style":329},[22009],{"type":26,"value":3738},{"type":21,"tag":322,"props":22011,"children":22012},{"class":324,"line":431},[22013,22018,22022,22027,22031,22035,22039],{"type":21,"tag":322,"props":22014,"children":22015},{"style":329},[22016],{"type":26,"value":22017},"qc_qiskit.h(",{"type":21,"tag":322,"props":22019,"children":22020},{"style":348},[22021],{"type":26,"value":5605},{"type":21,"tag":322,"props":22023,"children":22024},{"style":329},[22025],{"type":26,"value":22026},"); qc_qiskit.cx(",{"type":21,"tag":322,"props":22028,"children":22029},{"style":348},[22030],{"type":26,"value":5605},{"type":21,"tag":322,"props":22032,"children":22033},{"style":329},[22034],{"type":26,"value":367},{"type":21,"tag":322,"props":22036,"children":22037},{"style":348},[22038],{"type":26,"value":3994},{"type":21,"tag":322,"props":22040,"children":22041},{"style":329},[22042],{"type":26,"value":22043},"); qc_qiskit.measure_all()\n",{"type":21,"tag":322,"props":22045,"children":22046},{"class":324,"line":453},[22047,22052,22056,22061,22065,22069,22073],{"type":21,"tag":322,"props":22048,"children":22049},{"style":329},[22050],{"type":26,"value":22051},"result_qiskit ",{"type":21,"tag":322,"props":22053,"children":22054},{"style":3638},[22055],{"type":26,"value":3678},{"type":21,"tag":322,"props":22057,"children":22058},{"style":329},[22059],{"type":26,"value":22060}," AerSimulator().run(qc_qiskit, ",{"type":21,"tag":322,"props":22062,"children":22063},{"style":3703},[22064],{"type":26,"value":4560},{"type":21,"tag":322,"props":22066,"children":22067},{"style":3638},[22068],{"type":26,"value":3678},{"type":21,"tag":322,"props":22070,"children":22071},{"style":348},[22072],{"type":26,"value":20897},{"type":21,"tag":322,"props":22074,"children":22075},{"style":329},[22076],{"type":26,"value":20902},{"type":21,"tag":322,"props":22078,"children":22079},{"class":324,"line":474},[22080],{"type":21,"tag":322,"props":22081,"children":22082},{"emptyLinePlaceholder":3662},[22083],{"type":26,"value":3665},{"type":21,"tag":322,"props":22085,"children":22086},{"class":324,"line":496},[22087],{"type":21,"tag":322,"props":22088,"children":22089},{"style":3888},[22090],{"type":26,"value":22091},"# PennyLane — 全く異なるAPI、回路を書き直す\n",{"type":21,"tag":322,"props":22093,"children":22094},{"class":324,"line":16},[22095,22099,22103,22107],{"type":21,"tag":322,"props":22096,"children":22097},{"style":3638},[22098],{"type":26,"value":3651},{"type":21,"tag":322,"props":22100,"children":22101},{"style":329},[22102],{"type":26,"value":14530},{"type":21,"tag":322,"props":22104,"children":22105},{"style":3638},[22106],{"type":26,"value":4715},{"type":21,"tag":322,"props":22108,"children":22109},{"style":329},[22110],{"type":26,"value":14539},{"type":21,"tag":322,"props":22112,"children":22113},{"class":324,"line":3884},[22114,22118,22122,22126,22131,22135,22139,22143,22147],{"type":21,"tag":322,"props":22115,"children":22116},{"style":329},[22117],{"type":26,"value":14663},{"type":21,"tag":322,"props":22119,"children":22120},{"style":3638},[22121],{"type":26,"value":3678},{"type":21,"tag":322,"props":22123,"children":22124},{"style":329},[22125],{"type":26,"value":14672},{"type":21,"tag":322,"props":22127,"children":22128},{"style":359},[22129],{"type":26,"value":22130},"'default.qubit'",{"type":21,"tag":322,"props":22132,"children":22133},{"style":329},[22134],{"type":26,"value":367},{"type":21,"tag":322,"props":22136,"children":22137},{"style":3703},[22138],{"type":26,"value":14686},{"type":21,"tag":322,"props":22140,"children":22141},{"style":3638},[22142],{"type":26,"value":3678},{"type":21,"tag":322,"props":22144,"children":22145},{"style":348},[22146],{"type":26,"value":423},{"type":21,"tag":322,"props":22148,"children":22149},{"style":329},[22150],{"type":26,"value":3738},{"type":21,"tag":322,"props":22152,"children":22153},{"class":324,"line":3894},[22154,22158],{"type":21,"tag":322,"props":22155,"children":22156},{"style":4801},[22157],{"type":26,"value":14710},{"type":21,"tag":322,"props":22159,"children":22160},{"style":329},[22161],{"type":26,"value":14715},{"type":21,"tag":322,"props":22163,"children":22164},{"class":324,"line":1034},[22165,22169,22174],{"type":21,"tag":322,"props":22166,"children":22167},{"style":3638},[22168],{"type":26,"value":4798},{"type":21,"tag":322,"props":22170,"children":22171},{"style":4801},[22172],{"type":26,"value":22173}," bell",{"type":21,"tag":322,"props":22175,"children":22176},{"style":329},[22177],{"type":26,"value":22178},"():\n",{"type":21,"tag":322,"props":22180,"children":22181},{"class":324,"line":3954},[22182,22187,22191,22195,22199],{"type":21,"tag":322,"props":22183,"children":22184},{"style":329},[22185],{"type":26,"value":22186},"    qml.Hadamard(",{"type":21,"tag":322,"props":22188,"children":22189},{"style":3703},[22190],{"type":26,"value":14686},{"type":21,"tag":322,"props":22192,"children":22193},{"style":3638},[22194],{"type":26,"value":3678},{"type":21,"tag":322,"props":22196,"children":22197},{"style":348},[22198],{"type":26,"value":5605},{"type":21,"tag":322,"props":22200,"children":22201},{"style":329},[22202],{"type":26,"value":3738},{"type":21,"tag":322,"props":22204,"children":22205},{"class":324,"line":4002},[22206,22211,22215,22219,22224,22228,22232,22236],{"type":21,"tag":322,"props":22207,"children":22208},{"style":329},[22209],{"type":26,"value":22210},"    qml.CNOT(",{"type":21,"tag":322,"props":22212,"children":22213},{"style":3703},[22214],{"type":26,"value":14686},{"type":21,"tag":322,"props":22216,"children":22217},{"style":3638},[22218],{"type":26,"value":3678},{"type":21,"tag":322,"props":22220,"children":22221},{"style":329},[22222],{"type":26,"value":22223},"[",{"type":21,"tag":322,"props":22225,"children":22226},{"style":348},[22227],{"type":26,"value":5605},{"type":21,"tag":322,"props":22229,"children":22230},{"style":329},[22231],{"type":26,"value":367},{"type":21,"tag":322,"props":22233,"children":22234},{"style":348},[22235],{"type":26,"value":3994},{"type":21,"tag":322,"props":22237,"children":22238},{"style":329},[22239],{"type":26,"value":3999},{"type":21,"tag":322,"props":22241,"children":22242},{"class":324,"line":4044},[22243,22247,22252,22256,22260,22264,22268,22272,22276],{"type":21,"tag":322,"props":22244,"children":22245},{"style":3638},[22246],{"type":26,"value":5023},{"type":21,"tag":322,"props":22248,"children":22249},{"style":329},[22250],{"type":26,"value":22251}," qml.probs(",{"type":21,"tag":322,"props":22253,"children":22254},{"style":3703},[22255],{"type":26,"value":14686},{"type":21,"tag":322,"props":22257,"children":22258},{"style":3638},[22259],{"type":26,"value":3678},{"type":21,"tag":322,"props":22261,"children":22262},{"style":329},[22263],{"type":26,"value":22223},{"type":21,"tag":322,"props":22265,"children":22266},{"style":348},[22267],{"type":26,"value":5605},{"type":21,"tag":322,"props":22269,"children":22270},{"style":329},[22271],{"type":26,"value":367},{"type":21,"tag":322,"props":22273,"children":22274},{"style":348},[22275],{"type":26,"value":3994},{"type":21,"tag":322,"props":22277,"children":22278},{"style":329},[22279],{"type":26,"value":3999},{"type":21,"tag":322,"props":22281,"children":22282},{"class":324,"line":5068},[22283,22288,22292],{"type":21,"tag":322,"props":22284,"children":22285},{"style":329},[22286],{"type":26,"value":22287},"result_pennylane ",{"type":21,"tag":322,"props":22289,"children":22290},{"style":3638},[22291],{"type":26,"value":3678},{"type":21,"tag":322,"props":22293,"children":22294},{"style":329},[22295],{"type":26,"value":22296}," bell()\n",{"type":21,"tag":322,"props":22298,"children":22299},{"class":324,"line":5087},[22300],{"type":21,"tag":322,"props":22301,"children":22302},{"emptyLinePlaceholder":3662},[22303],{"type":26,"value":3665},{"type":21,"tag":322,"props":22305,"children":22306},{"class":324,"line":5095},[22307],{"type":21,"tag":322,"props":22308,"children":22309},{"style":3888},[22310],{"type":26,"value":22311},"# CUDA-Q — さらに別のAPI、また書き直す\n",{"type":21,"tag":322,"props":22313,"children":22314},{"class":324,"line":5126},[22315,22319],{"type":21,"tag":322,"props":22316,"children":22317},{"style":3638},[22318],{"type":26,"value":3651},{"type":21,"tag":322,"props":22320,"children":22321},{"style":329},[22322],{"type":26,"value":22323}," cudaq\n",{"type":21,"tag":322,"props":22325,"children":22326},{"class":324,"line":5171},[22327],{"type":21,"tag":322,"props":22328,"children":22329},{"style":4801},[22330],{"type":26,"value":22331},"@cudaq.kernel\n",{"type":21,"tag":322,"props":22333,"children":22334},{"class":324,"line":5210},[22335,22339,22343],{"type":21,"tag":322,"props":22336,"children":22337},{"style":3638},[22338],{"type":26,"value":4798},{"type":21,"tag":322,"props":22340,"children":22341},{"style":4801},[22342],{"type":26,"value":22173},{"type":21,"tag":322,"props":22344,"children":22345},{"style":329},[22346],{"type":26,"value":22178},{"type":21,"tag":322,"props":22348,"children":22349},{"class":324,"line":5228},[22350,22355,22359,22364,22368],{"type":21,"tag":322,"props":22351,"children":22352},{"style":329},[22353],{"type":26,"value":22354},"    qvec ",{"type":21,"tag":322,"props":22356,"children":22357},{"style":3638},[22358],{"type":26,"value":3678},{"type":21,"tag":322,"props":22360,"children":22361},{"style":329},[22362],{"type":26,"value":22363}," cudaq.qvector(",{"type":21,"tag":322,"props":22365,"children":22366},{"style":348},[22367],{"type":26,"value":423},{"type":21,"tag":322,"props":22369,"children":22370},{"style":329},[22371],{"type":26,"value":3738},{"type":21,"tag":322,"props":22373,"children":22374},{"class":324,"line":5302},[22375,22380,22384,22389,22393,22398,22402],{"type":21,"tag":322,"props":22376,"children":22377},{"style":329},[22378],{"type":26,"value":22379},"    h(qvec[",{"type":21,"tag":322,"props":22381,"children":22382},{"style":348},[22383],{"type":26,"value":5605},{"type":21,"tag":322,"props":22385,"children":22386},{"style":329},[22387],{"type":26,"value":22388},"]); cx(qvec[",{"type":21,"tag":322,"props":22390,"children":22391},{"style":348},[22392],{"type":26,"value":5605},{"type":21,"tag":322,"props":22394,"children":22395},{"style":329},[22396],{"type":26,"value":22397},"], qvec[",{"type":21,"tag":322,"props":22399,"children":22400},{"style":348},[22401],{"type":26,"value":3994},{"type":21,"tag":322,"props":22403,"children":22404},{"style":329},[22405],{"type":26,"value":22406},"]); mz(qvec)\n",{"type":21,"tag":322,"props":22408,"children":22409},{"class":324,"line":5330},[22410,22415,22419,22424,22429,22433,22437],{"type":21,"tag":322,"props":22411,"children":22412},{"style":329},[22413],{"type":26,"value":22414},"result_cudaq ",{"type":21,"tag":322,"props":22416,"children":22417},{"style":3638},[22418],{"type":26,"value":3678},{"type":21,"tag":322,"props":22420,"children":22421},{"style":329},[22422],{"type":26,"value":22423}," cudaq.sample(bell, ",{"type":21,"tag":322,"props":22425,"children":22426},{"style":3703},[22427],{"type":26,"value":22428},"shots_count",{"type":21,"tag":322,"props":22430,"children":22431},{"style":3638},[22432],{"type":26,"value":3678},{"type":21,"tag":322,"props":22434,"children":22435},{"style":348},[22436],{"type":26,"value":20897},{"type":21,"tag":322,"props":22438,"children":22439},{"style":329},[22440],{"type":26,"value":3738},{"type":21,"tag":22,"props":22442,"children":22443},{},[22444],{"type":21,"tag":71,"props":22445,"children":22446},{},[22447],{"type":26,"value":22448},"HLQuantum(1つの回路、任意のバックエンド):",{"type":21,"tag":312,"props":22450,"children":22452},{"code":22451,"language":3628,"meta":7,"className":3626,"style":7},"import hlquantum as hlq\n\nqc = hlq.Circuit(2)\nqc.h(0).cx(0, 1).measure_all()\n\n# 一語でバックエンドを切り替え\nresult_qiskit    = hlq.run(qc, backend='qiskit')\nresult_pennylane = hlq.run(qc, backend='pennylane')\nresult_cudaq     = hlq.run(qc, backend='cudaq')    # GPU\nresult_cirq      = hlq.run(qc, backend='cirq')\nresult_ionq      = hlq.run(qc, backend='ionq', device='aria-1')\nresult_ibm       = hlq.run(qc, backend='qiskit', device='ibm_sherbrooke')\n",[22453],{"type":21,"tag":79,"props":22454,"children":22455},{"__ignoreMap":7},[22456,22475,22482,22505,22536,22543,22551,22583,22615,22653,22686,22737],{"type":21,"tag":322,"props":22457,"children":22458},{"class":324,"line":325},[22459,22463,22467,22471],{"type":21,"tag":322,"props":22460,"children":22461},{"style":3638},[22462],{"type":26,"value":3651},{"type":21,"tag":322,"props":22464,"children":22465},{"style":329},[22466],{"type":26,"value":19732},{"type":21,"tag":322,"props":22468,"children":22469},{"style":3638},[22470],{"type":26,"value":4715},{"type":21,"tag":322,"props":22472,"children":22473},{"style":329},[22474],{"type":26,"value":19741},{"type":21,"tag":322,"props":22476,"children":22477},{"class":324,"line":335},[22478],{"type":21,"tag":322,"props":22479,"children":22480},{"emptyLinePlaceholder":3662},[22481],{"type":26,"value":3665},{"type":21,"tag":322,"props":22483,"children":22484},{"class":324,"line":344},[22485,22489,22493,22497,22501],{"type":21,"tag":322,"props":22486,"children":22487},{"style":329},[22488],{"type":26,"value":7819},{"type":21,"tag":322,"props":22490,"children":22491},{"style":3638},[22492],{"type":26,"value":3678},{"type":21,"tag":322,"props":22494,"children":22495},{"style":329},[22496],{"type":26,"value":21793},{"type":21,"tag":322,"props":22498,"children":22499},{"style":348},[22500],{"type":26,"value":423},{"type":21,"tag":322,"props":22502,"children":22503},{"style":329},[22504],{"type":26,"value":3738},{"type":21,"tag":322,"props":22506,"children":22507},{"class":324,"line":407},[22508,22512,22516,22520,22524,22528,22532],{"type":21,"tag":322,"props":22509,"children":22510},{"style":329},[22511],{"type":26,"value":7835},{"type":21,"tag":322,"props":22513,"children":22514},{"style":348},[22515],{"type":26,"value":5605},{"type":21,"tag":322,"props":22517,"children":22518},{"style":329},[22519],{"type":26,"value":21817},{"type":21,"tag":322,"props":22521,"children":22522},{"style":348},[22523],{"type":26,"value":5605},{"type":21,"tag":322,"props":22525,"children":22526},{"style":329},[22527],{"type":26,"value":367},{"type":21,"tag":322,"props":22529,"children":22530},{"style":348},[22531],{"type":26,"value":3994},{"type":21,"tag":322,"props":22533,"children":22534},{"style":329},[22535],{"type":26,"value":21834},{"type":21,"tag":322,"props":22537,"children":22538},{"class":324,"line":431},[22539],{"type":21,"tag":322,"props":22540,"children":22541},{"emptyLinePlaceholder":3662},[22542],{"type":26,"value":3665},{"type":21,"tag":322,"props":22544,"children":22545},{"class":324,"line":453},[22546],{"type":21,"tag":322,"props":22547,"children":22548},{"style":3888},[22549],{"type":26,"value":22550},"# 一語でバックエンドを切り替え\n",{"type":21,"tag":322,"props":22552,"children":22553},{"class":324,"line":474},[22554,22559,22563,22567,22571,22575,22579],{"type":21,"tag":322,"props":22555,"children":22556},{"style":329},[22557],{"type":26,"value":22558},"result_qiskit    ",{"type":21,"tag":322,"props":22560,"children":22561},{"style":3638},[22562],{"type":26,"value":3678},{"type":21,"tag":322,"props":22564,"children":22565},{"style":329},[22566],{"type":26,"value":21857},{"type":21,"tag":322,"props":22568,"children":22569},{"style":3703},[22570],{"type":26,"value":18908},{"type":21,"tag":322,"props":22572,"children":22573},{"style":3638},[22574],{"type":26,"value":3678},{"type":21,"tag":322,"props":22576,"children":22577},{"style":359},[22578],{"type":26,"value":19822},{"type":21,"tag":322,"props":22580,"children":22581},{"style":329},[22582],{"type":26,"value":3738},{"type":21,"tag":322,"props":22584,"children":22585},{"class":324,"line":496},[22586,22590,22594,22598,22602,22606,22611],{"type":21,"tag":322,"props":22587,"children":22588},{"style":329},[22589],{"type":26,"value":22287},{"type":21,"tag":322,"props":22591,"children":22592},{"style":3638},[22593],{"type":26,"value":3678},{"type":21,"tag":322,"props":22595,"children":22596},{"style":329},[22597],{"type":26,"value":21857},{"type":21,"tag":322,"props":22599,"children":22600},{"style":3703},[22601],{"type":26,"value":18908},{"type":21,"tag":322,"props":22603,"children":22604},{"style":3638},[22605],{"type":26,"value":3678},{"type":21,"tag":322,"props":22607,"children":22608},{"style":359},[22609],{"type":26,"value":22610},"'pennylane'",{"type":21,"tag":322,"props":22612,"children":22613},{"style":329},[22614],{"type":26,"value":3738},{"type":21,"tag":322,"props":22616,"children":22617},{"class":324,"line":16},[22618,22623,22627,22631,22635,22639,22644,22648],{"type":21,"tag":322,"props":22619,"children":22620},{"style":329},[22621],{"type":26,"value":22622},"result_cudaq     ",{"type":21,"tag":322,"props":22624,"children":22625},{"style":3638},[22626],{"type":26,"value":3678},{"type":21,"tag":322,"props":22628,"children":22629},{"style":329},[22630],{"type":26,"value":21857},{"type":21,"tag":322,"props":22632,"children":22633},{"style":3703},[22634],{"type":26,"value":18908},{"type":21,"tag":322,"props":22636,"children":22637},{"style":3638},[22638],{"type":26,"value":3678},{"type":21,"tag":322,"props":22640,"children":22641},{"style":359},[22642],{"type":26,"value":22643},"'cudaq'",{"type":21,"tag":322,"props":22645,"children":22646},{"style":329},[22647],{"type":26,"value":8649},{"type":21,"tag":322,"props":22649,"children":22650},{"style":3888},[22651],{"type":26,"value":22652},"# GPU\n",{"type":21,"tag":322,"props":22654,"children":22655},{"class":324,"line":3884},[22656,22661,22665,22669,22673,22677,22682],{"type":21,"tag":322,"props":22657,"children":22658},{"style":329},[22659],{"type":26,"value":22660},"result_cirq      ",{"type":21,"tag":322,"props":22662,"children":22663},{"style":3638},[22664],{"type":26,"value":3678},{"type":21,"tag":322,"props":22666,"children":22667},{"style":329},[22668],{"type":26,"value":21857},{"type":21,"tag":322,"props":22670,"children":22671},{"style":3703},[22672],{"type":26,"value":18908},{"type":21,"tag":322,"props":22674,"children":22675},{"style":3638},[22676],{"type":26,"value":3678},{"type":21,"tag":322,"props":22678,"children":22679},{"style":359},[22680],{"type":26,"value":22681},"'cirq'",{"type":21,"tag":322,"props":22683,"children":22684},{"style":329},[22685],{"type":26,"value":3738},{"type":21,"tag":322,"props":22687,"children":22688},{"class":324,"line":3894},[22689,22694,22698,22702,22706,22710,22715,22719,22724,22728,22733],{"type":21,"tag":322,"props":22690,"children":22691},{"style":329},[22692],{"type":26,"value":22693},"result_ionq      ",{"type":21,"tag":322,"props":22695,"children":22696},{"style":3638},[22697],{"type":26,"value":3678},{"type":21,"tag":322,"props":22699,"children":22700},{"style":329},[22701],{"type":26,"value":21857},{"type":21,"tag":322,"props":22703,"children":22704},{"style":3703},[22705],{"type":26,"value":18908},{"type":21,"tag":322,"props":22707,"children":22708},{"style":3638},[22709],{"type":26,"value":3678},{"type":21,"tag":322,"props":22711,"children":22712},{"style":359},[22713],{"type":26,"value":22714},"'ionq'",{"type":21,"tag":322,"props":22716,"children":22717},{"style":329},[22718],{"type":26,"value":367},{"type":21,"tag":322,"props":22720,"children":22721},{"style":3703},[22722],{"type":26,"value":22723},"device",{"type":21,"tag":322,"props":22725,"children":22726},{"style":3638},[22727],{"type":26,"value":3678},{"type":21,"tag":322,"props":22729,"children":22730},{"style":359},[22731],{"type":26,"value":22732},"'aria-1'",{"type":21,"tag":322,"props":22734,"children":22735},{"style":329},[22736],{"type":26,"value":3738},{"type":21,"tag":322,"props":22738,"children":22739},{"class":324,"line":1034},[22740,22745,22749,22753,22757,22761,22765,22769,22773,22777,22782],{"type":21,"tag":322,"props":22741,"children":22742},{"style":329},[22743],{"type":26,"value":22744},"result_ibm       ",{"type":21,"tag":322,"props":22746,"children":22747},{"style":3638},[22748],{"type":26,"value":3678},{"type":21,"tag":322,"props":22750,"children":22751},{"style":329},[22752],{"type":26,"value":21857},{"type":21,"tag":322,"props":22754,"children":22755},{"style":3703},[22756],{"type":26,"value":18908},{"type":21,"tag":322,"props":22758,"children":22759},{"style":3638},[22760],{"type":26,"value":3678},{"type":21,"tag":322,"props":22762,"children":22763},{"style":359},[22764],{"type":26,"value":19822},{"type":21,"tag":322,"props":22766,"children":22767},{"style":329},[22768],{"type":26,"value":367},{"type":21,"tag":322,"props":22770,"children":22771},{"style":3703},[22772],{"type":26,"value":22723},{"type":21,"tag":322,"props":22774,"children":22775},{"style":3638},[22776],{"type":26,"value":3678},{"type":21,"tag":322,"props":22778,"children":22779},{"style":359},[22780],{"type":26,"value":22781},"'ibm_sherbrooke'",{"type":21,"tag":322,"props":22783,"children":22784},{"style":329},[22785],{"type":26,"value":3738},{"type":21,"tag":22,"props":22787,"children":22788},{},[22789],{"type":26,"value":22790},"ここでHLQuantumの真価が発揮されます。ローカルシミュレータでプロトタイプを作り、回路を書き直すことなくQPUへデプロイできるのです。",{"type":21,"tag":59,"props":22792,"children":22794},{"id":22793},"並列比較-vqe",[22795],{"type":26,"value":22796},"並列比較: VQE",{"type":21,"tag":22,"props":22798,"children":22799},{},[22800],{"type":21,"tag":71,"props":22801,"children":22802},{},[22803],{"type":26,"value":22804},"素のQiskit VQE:",{"type":21,"tag":312,"props":22806,"children":22808},{"code":22807,"language":3628,"meta":7,"className":3626,"style":7},"from qiskit_nature.second_q.drivers import PySCFDriver\nfrom qiskit_nature.second_q.mappers import JordanWignerMapper\nfrom qiskit_nature.second_q.circuit.library import UCCSD\nfrom qiskit_algorithms import VQE\nfrom qiskit_algorithms.optimizers import COBYLA\nfrom qiskit.primitives import Estimator\n\ndriver = PySCFDriver(atom='H .0 .0 .0; H .0 .0 0.735')\nproblem = driver.run()\nmapper = JordanWignerMapper()\nhamiltonian = mapper.map(problem.second_q_ops()[0])\n\nansatz = UCCSD(\n    problem.num_spatial_orbitals,\n    problem.num_particles,\n    mapper,\n    initial_state=problem.reference_state,\n)\n\nvqe = VQE(Estimator(), ansatz, COBYLA())\nresult = vqe.compute_minimum_eigenvalue(hamiltonian)\nprint(f\"H2 ground state: {result.eigenvalue.real:.4f} Ha\")\n",[22809],{"type":21,"tag":79,"props":22810,"children":22811},{"__ignoreMap":7},[22812,22833,22854,22875,22896,22917,22938,22945,22980,22997,23014,23039,23046,23063,23071,23079,23087,23104,23111,23118,23135,23151],{"type":21,"tag":322,"props":22813,"children":22814},{"class":324,"line":325},[22815,22819,22824,22828],{"type":21,"tag":322,"props":22816,"children":22817},{"style":3638},[22818],{"type":26,"value":3641},{"type":21,"tag":322,"props":22820,"children":22821},{"style":329},[22822],{"type":26,"value":22823}," qiskit_nature.second_q.drivers ",{"type":21,"tag":322,"props":22825,"children":22826},{"style":3638},[22827],{"type":26,"value":3651},{"type":21,"tag":322,"props":22829,"children":22830},{"style":329},[22831],{"type":26,"value":22832}," PySCFDriver\n",{"type":21,"tag":322,"props":22834,"children":22835},{"class":324,"line":335},[22836,22840,22845,22849],{"type":21,"tag":322,"props":22837,"children":22838},{"style":3638},[22839],{"type":26,"value":3641},{"type":21,"tag":322,"props":22841,"children":22842},{"style":329},[22843],{"type":26,"value":22844}," qiskit_nature.second_q.mappers ",{"type":21,"tag":322,"props":22846,"children":22847},{"style":3638},[22848],{"type":26,"value":3651},{"type":21,"tag":322,"props":22850,"children":22851},{"style":329},[22852],{"type":26,"value":22853}," JordanWignerMapper\n",{"type":21,"tag":322,"props":22855,"children":22856},{"class":324,"line":344},[22857,22861,22866,22870],{"type":21,"tag":322,"props":22858,"children":22859},{"style":3638},[22860],{"type":26,"value":3641},{"type":21,"tag":322,"props":22862,"children":22863},{"style":329},[22864],{"type":26,"value":22865}," qiskit_nature.second_q.circuit.library ",{"type":21,"tag":322,"props":22867,"children":22868},{"style":3638},[22869],{"type":26,"value":3651},{"type":21,"tag":322,"props":22871,"children":22872},{"style":348},[22873],{"type":26,"value":22874}," UCCSD\n",{"type":21,"tag":322,"props":22876,"children":22877},{"class":324,"line":407},[22878,22882,22887,22891],{"type":21,"tag":322,"props":22879,"children":22880},{"style":3638},[22881],{"type":26,"value":3641},{"type":21,"tag":322,"props":22883,"children":22884},{"style":329},[22885],{"type":26,"value":22886}," qiskit_algorithms ",{"type":21,"tag":322,"props":22888,"children":22889},{"style":3638},[22890],{"type":26,"value":3651},{"type":21,"tag":322,"props":22892,"children":22893},{"style":348},[22894],{"type":26,"value":22895}," VQE\n",{"type":21,"tag":322,"props":22897,"children":22898},{"class":324,"line":431},[22899,22903,22908,22912],{"type":21,"tag":322,"props":22900,"children":22901},{"style":3638},[22902],{"type":26,"value":3641},{"type":21,"tag":322,"props":22904,"children":22905},{"style":329},[22906],{"type":26,"value":22907}," qiskit_algorithms.optimizers ",{"type":21,"tag":322,"props":22909,"children":22910},{"style":3638},[22911],{"type":26,"value":3651},{"type":21,"tag":322,"props":22913,"children":22914},{"style":348},[22915],{"type":26,"value":22916}," COBYLA\n",{"type":21,"tag":322,"props":22918,"children":22919},{"class":324,"line":453},[22920,22924,22929,22933],{"type":21,"tag":322,"props":22921,"children":22922},{"style":3638},[22923],{"type":26,"value":3641},{"type":21,"tag":322,"props":22925,"children":22926},{"style":329},[22927],{"type":26,"value":22928}," qiskit.primitives ",{"type":21,"tag":322,"props":22930,"children":22931},{"style":3638},[22932],{"type":26,"value":3651},{"type":21,"tag":322,"props":22934,"children":22935},{"style":329},[22936],{"type":26,"value":22937}," Estimator\n",{"type":21,"tag":322,"props":22939,"children":22940},{"class":324,"line":474},[22941],{"type":21,"tag":322,"props":22942,"children":22943},{"emptyLinePlaceholder":3662},[22944],{"type":26,"value":3665},{"type":21,"tag":322,"props":22946,"children":22947},{"class":324,"line":496},[22948,22953,22957,22962,22967,22971,22976],{"type":21,"tag":322,"props":22949,"children":22950},{"style":329},[22951],{"type":26,"value":22952},"driver ",{"type":21,"tag":322,"props":22954,"children":22955},{"style":3638},[22956],{"type":26,"value":3678},{"type":21,"tag":322,"props":22958,"children":22959},{"style":329},[22960],{"type":26,"value":22961}," PySCFDriver(",{"type":21,"tag":322,"props":22963,"children":22964},{"style":3703},[22965],{"type":26,"value":22966},"atom",{"type":21,"tag":322,"props":22968,"children":22969},{"style":3638},[22970],{"type":26,"value":3678},{"type":21,"tag":322,"props":22972,"children":22973},{"style":359},[22974],{"type":26,"value":22975},"'H .0 .0 .0; H .0 .0 0.735'",{"type":21,"tag":322,"props":22977,"children":22978},{"style":329},[22979],{"type":26,"value":3738},{"type":21,"tag":322,"props":22981,"children":22982},{"class":324,"line":16},[22983,22988,22992],{"type":21,"tag":322,"props":22984,"children":22985},{"style":329},[22986],{"type":26,"value":22987},"problem ",{"type":21,"tag":322,"props":22989,"children":22990},{"style":3638},[22991],{"type":26,"value":3678},{"type":21,"tag":322,"props":22993,"children":22994},{"style":329},[22995],{"type":26,"value":22996}," driver.run()\n",{"type":21,"tag":322,"props":22998,"children":22999},{"class":324,"line":3884},[23000,23005,23009],{"type":21,"tag":322,"props":23001,"children":23002},{"style":329},[23003],{"type":26,"value":23004},"mapper ",{"type":21,"tag":322,"props":23006,"children":23007},{"style":3638},[23008],{"type":26,"value":3678},{"type":21,"tag":322,"props":23010,"children":23011},{"style":329},[23012],{"type":26,"value":23013}," JordanWignerMapper()\n",{"type":21,"tag":322,"props":23015,"children":23016},{"class":324,"line":3894},[23017,23022,23026,23031,23035],{"type":21,"tag":322,"props":23018,"children":23019},{"style":329},[23020],{"type":26,"value":23021},"hamiltonian ",{"type":21,"tag":322,"props":23023,"children":23024},{"style":3638},[23025],{"type":26,"value":3678},{"type":21,"tag":322,"props":23027,"children":23028},{"style":329},[23029],{"type":26,"value":23030}," mapper.map(problem.second_q_ops()[",{"type":21,"tag":322,"props":23032,"children":23033},{"style":348},[23034],{"type":26,"value":5605},{"type":21,"tag":322,"props":23036,"children":23037},{"style":329},[23038],{"type":26,"value":3999},{"type":21,"tag":322,"props":23040,"children":23041},{"class":324,"line":1034},[23042],{"type":21,"tag":322,"props":23043,"children":23044},{"emptyLinePlaceholder":3662},[23045],{"type":26,"value":3665},{"type":21,"tag":322,"props":23047,"children":23048},{"class":324,"line":3954},[23049,23054,23058],{"type":21,"tag":322,"props":23050,"children":23051},{"style":329},[23052],{"type":26,"value":23053},"ansatz ",{"type":21,"tag":322,"props":23055,"children":23056},{"style":3638},[23057],{"type":26,"value":3678},{"type":21,"tag":322,"props":23059,"children":23060},{"style":329},[23061],{"type":26,"value":23062}," UCCSD(\n",{"type":21,"tag":322,"props":23064,"children":23065},{"class":324,"line":4002},[23066],{"type":21,"tag":322,"props":23067,"children":23068},{"style":329},[23069],{"type":26,"value":23070},"    problem.num_spatial_orbitals,\n",{"type":21,"tag":322,"props":23072,"children":23073},{"class":324,"line":4044},[23074],{"type":21,"tag":322,"props":23075,"children":23076},{"style":329},[23077],{"type":26,"value":23078},"    problem.num_particles,\n",{"type":21,"tag":322,"props":23080,"children":23081},{"class":324,"line":5068},[23082],{"type":21,"tag":322,"props":23083,"children":23084},{"style":329},[23085],{"type":26,"value":23086},"    mapper,\n",{"type":21,"tag":322,"props":23088,"children":23089},{"class":324,"line":5087},[23090,23095,23099],{"type":21,"tag":322,"props":23091,"children":23092},{"style":3703},[23093],{"type":26,"value":23094},"    initial_state",{"type":21,"tag":322,"props":23096,"children":23097},{"style":3638},[23098],{"type":26,"value":3678},{"type":21,"tag":322,"props":23100,"children":23101},{"style":329},[23102],{"type":26,"value":23103},"problem.reference_state,\n",{"type":21,"tag":322,"props":23105,"children":23106},{"class":324,"line":5095},[23107],{"type":21,"tag":322,"props":23108,"children":23109},{"style":329},[23110],{"type":26,"value":3738},{"type":21,"tag":322,"props":23112,"children":23113},{"class":324,"line":5126},[23114],{"type":21,"tag":322,"props":23115,"children":23116},{"emptyLinePlaceholder":3662},[23117],{"type":26,"value":3665},{"type":21,"tag":322,"props":23119,"children":23120},{"class":324,"line":5171},[23121,23126,23130],{"type":21,"tag":322,"props":23122,"children":23123},{"style":329},[23124],{"type":26,"value":23125},"vqe ",{"type":21,"tag":322,"props":23127,"children":23128},{"style":3638},[23129],{"type":26,"value":3678},{"type":21,"tag":322,"props":23131,"children":23132},{"style":329},[23133],{"type":26,"value":23134}," VQE(Estimator(), ansatz, COBYLA())\n",{"type":21,"tag":322,"props":23136,"children":23137},{"class":324,"line":5210},[23138,23142,23146],{"type":21,"tag":322,"props":23139,"children":23140},{"style":329},[23141],{"type":26,"value":19764},{"type":21,"tag":322,"props":23143,"children":23144},{"style":3638},[23145],{"type":26,"value":3678},{"type":21,"tag":322,"props":23147,"children":23148},{"style":329},[23149],{"type":26,"value":23150}," vqe.compute_minimum_eigenvalue(hamiltonian)\n",{"type":21,"tag":322,"props":23152,"children":23153},{"class":324,"line":5228},[23154,23158,23162,23166,23171,23175,23180,23184,23188,23192],{"type":21,"tag":322,"props":23155,"children":23156},{"style":348},[23157],{"type":26,"value":3753},{"type":21,"tag":322,"props":23159,"children":23160},{"style":329},[23161],{"type":26,"value":3758},{"type":21,"tag":322,"props":23163,"children":23164},{"style":3638},[23165],{"type":26,"value":3763},{"type":21,"tag":322,"props":23167,"children":23168},{"style":359},[23169],{"type":26,"value":23170},"\"H2 ground state: ",{"type":21,"tag":322,"props":23172,"children":23173},{"style":348},[23174],{"type":26,"value":3773},{"type":21,"tag":322,"props":23176,"children":23177},{"style":329},[23178],{"type":26,"value":23179},"result.eigenvalue.real",{"type":21,"tag":322,"props":23181,"children":23182},{"style":3638},[23183],{"type":26,"value":4085},{"type":21,"tag":322,"props":23185,"children":23186},{"style":348},[23187],{"type":26,"value":3783},{"type":21,"tag":322,"props":23189,"children":23190},{"style":359},[23191],{"type":26,"value":19868},{"type":21,"tag":322,"props":23193,"children":23194},{"style":329},[23195],{"type":26,"value":3738},{"type":21,"tag":22,"props":23197,"children":23198},{},[23199],{"type":21,"tag":71,"props":23200,"children":23201},{},[23202],{"type":26,"value":23203},"HLQuantum VQE:",{"type":21,"tag":312,"props":23205,"children":23207},{"code":23206,"language":3628,"meta":7,"className":3626,"style":7},"import hlquantum as hlq\n\nresult = hlq.algorithms.vqe(\n    molecule='H2',\n    basis='sto-3g',\n    backend='qiskit',\n    optimizer='COBYLA',\n)\nprint(f\"H2 ground state: {result.energy:.4f} Ha\")\n\n# GPUへ切り替え:\nresult_gpu = hlq.algorithms.vqe(molecule='H2', backend='cudaq')\n",[23208],{"type":21,"tag":79,"props":23209,"children":23210},{"__ignoreMap":7},[23211,23230,23237,23253,23273,23293,23313,23334,23341,23384,23391,23399],{"type":21,"tag":322,"props":23212,"children":23213},{"class":324,"line":325},[23214,23218,23222,23226],{"type":21,"tag":322,"props":23215,"children":23216},{"style":3638},[23217],{"type":26,"value":3651},{"type":21,"tag":322,"props":23219,"children":23220},{"style":329},[23221],{"type":26,"value":19732},{"type":21,"tag":322,"props":23223,"children":23224},{"style":3638},[23225],{"type":26,"value":4715},{"type":21,"tag":322,"props":23227,"children":23228},{"style":329},[23229],{"type":26,"value":19741},{"type":21,"tag":322,"props":23231,"children":23232},{"class":324,"line":335},[23233],{"type":21,"tag":322,"props":23234,"children":23235},{"emptyLinePlaceholder":3662},[23236],{"type":26,"value":3665},{"type":21,"tag":322,"props":23238,"children":23239},{"class":324,"line":344},[23240,23244,23248],{"type":21,"tag":322,"props":23241,"children":23242},{"style":329},[23243],{"type":26,"value":19764},{"type":21,"tag":322,"props":23245,"children":23246},{"style":3638},[23247],{"type":26,"value":3678},{"type":21,"tag":322,"props":23249,"children":23250},{"style":329},[23251],{"type":26,"value":23252}," hlq.algorithms.vqe(\n",{"type":21,"tag":322,"props":23254,"children":23255},{"class":324,"line":407},[23256,23261,23265,23269],{"type":21,"tag":322,"props":23257,"children":23258},{"style":3703},[23259],{"type":26,"value":23260},"    molecule",{"type":21,"tag":322,"props":23262,"children":23263},{"style":3638},[23264],{"type":26,"value":3678},{"type":21,"tag":322,"props":23266,"children":23267},{"style":359},[23268],{"type":26,"value":19787},{"type":21,"tag":322,"props":23270,"children":23271},{"style":329},[23272],{"type":26,"value":428},{"type":21,"tag":322,"props":23274,"children":23275},{"class":324,"line":431},[23276,23281,23285,23289],{"type":21,"tag":322,"props":23277,"children":23278},{"style":3703},[23279],{"type":26,"value":23280},"    basis",{"type":21,"tag":322,"props":23282,"children":23283},{"style":3638},[23284],{"type":26,"value":3678},{"type":21,"tag":322,"props":23286,"children":23287},{"style":359},[23288],{"type":26,"value":19805},{"type":21,"tag":322,"props":23290,"children":23291},{"style":329},[23292],{"type":26,"value":428},{"type":21,"tag":322,"props":23294,"children":23295},{"class":324,"line":453},[23296,23301,23305,23309],{"type":21,"tag":322,"props":23297,"children":23298},{"style":3703},[23299],{"type":26,"value":23300},"    backend",{"type":21,"tag":322,"props":23302,"children":23303},{"style":3638},[23304],{"type":26,"value":3678},{"type":21,"tag":322,"props":23306,"children":23307},{"style":359},[23308],{"type":26,"value":19822},{"type":21,"tag":322,"props":23310,"children":23311},{"style":329},[23312],{"type":26,"value":428},{"type":21,"tag":322,"props":23314,"children":23315},{"class":324,"line":474},[23316,23321,23325,23330],{"type":21,"tag":322,"props":23317,"children":23318},{"style":3703},[23319],{"type":26,"value":23320},"    optimizer",{"type":21,"tag":322,"props":23322,"children":23323},{"style":3638},[23324],{"type":26,"value":3678},{"type":21,"tag":322,"props":23326,"children":23327},{"style":359},[23328],{"type":26,"value":23329},"'COBYLA'",{"type":21,"tag":322,"props":23331,"children":23332},{"style":329},[23333],{"type":26,"value":428},{"type":21,"tag":322,"props":23335,"children":23336},{"class":324,"line":496},[23337],{"type":21,"tag":322,"props":23338,"children":23339},{"style":329},[23340],{"type":26,"value":3738},{"type":21,"tag":322,"props":23342,"children":23343},{"class":324,"line":16},[23344,23348,23352,23356,23360,23364,23368,23372,23376,23380],{"type":21,"tag":322,"props":23345,"children":23346},{"style":348},[23347],{"type":26,"value":3753},{"type":21,"tag":322,"props":23349,"children":23350},{"style":329},[23351],{"type":26,"value":3758},{"type":21,"tag":322,"props":23353,"children":23354},{"style":3638},[23355],{"type":26,"value":3763},{"type":21,"tag":322,"props":23357,"children":23358},{"style":359},[23359],{"type":26,"value":23170},{"type":21,"tag":322,"props":23361,"children":23362},{"style":348},[23363],{"type":26,"value":3773},{"type":21,"tag":322,"props":23365,"children":23366},{"style":329},[23367],{"type":26,"value":19855},{"type":21,"tag":322,"props":23369,"children":23370},{"style":3638},[23371],{"type":26,"value":4085},{"type":21,"tag":322,"props":23373,"children":23374},{"style":348},[23375],{"type":26,"value":3783},{"type":21,"tag":322,"props":23377,"children":23378},{"style":359},[23379],{"type":26,"value":19868},{"type":21,"tag":322,"props":23381,"children":23382},{"style":329},[23383],{"type":26,"value":3738},{"type":21,"tag":322,"props":23385,"children":23386},{"class":324,"line":3884},[23387],{"type":21,"tag":322,"props":23388,"children":23389},{"emptyLinePlaceholder":3662},[23390],{"type":26,"value":3665},{"type":21,"tag":322,"props":23392,"children":23393},{"class":324,"line":3894},[23394],{"type":21,"tag":322,"props":23395,"children":23396},{"style":3888},[23397],{"type":26,"value":23398},"# GPUへ切り替え:\n",{"type":21,"tag":322,"props":23400,"children":23401},{"class":324,"line":1034},[23402,23407,23411,23415,23419,23423,23427,23431,23435,23439,23443],{"type":21,"tag":322,"props":23403,"children":23404},{"style":329},[23405],{"type":26,"value":23406},"result_gpu ",{"type":21,"tag":322,"props":23408,"children":23409},{"style":3638},[23410],{"type":26,"value":3678},{"type":21,"tag":322,"props":23412,"children":23413},{"style":329},[23414],{"type":26,"value":19773},{"type":21,"tag":322,"props":23416,"children":23417},{"style":3703},[23418],{"type":26,"value":19778},{"type":21,"tag":322,"props":23420,"children":23421},{"style":3638},[23422],{"type":26,"value":3678},{"type":21,"tag":322,"props":23424,"children":23425},{"style":359},[23426],{"type":26,"value":19787},{"type":21,"tag":322,"props":23428,"children":23429},{"style":329},[23430],{"type":26,"value":367},{"type":21,"tag":322,"props":23432,"children":23433},{"style":3703},[23434],{"type":26,"value":18908},{"type":21,"tag":322,"props":23436,"children":23437},{"style":3638},[23438],{"type":26,"value":3678},{"type":21,"tag":322,"props":23440,"children":23441},{"style":359},[23442],{"type":26,"value":22643},{"type":21,"tag":322,"props":23444,"children":23445},{"style":329},[23446],{"type":26,"value":3738},{"type":21,"tag":22,"props":23448,"children":23449},{},[23450],{"type":26,"value":23451},"素のQiskitは、ansatz、mapper、driverに対する完全な制御を与えてくれます。HLQuantumのワンライナーは、バックエンドを比較したり、素早く結果を得たりするのに最適です。一方、カスタムansatzやアルゴリズムのチューニングが必要な場合は素のQiskitのほうが優れています。",{"type":21,"tag":59,"props":23453,"children":23455},{"id":23454},"素のqiskitを使うべきとき",[23456],{"type":26,"value":23457},"素のQiskitを使うべきとき",{"type":21,"tag":22,"props":23459,"children":23460},{},[23461],{"type":26,"value":23462},"Qiskitを直接使うべきなのは、次のような場合です。",{"type":21,"tag":7444,"props":23464,"children":23465},{},[23466,23476,23486,23525,23535],{"type":21,"tag":3560,"props":23467,"children":23468},{},[23469,23474],{"type":21,"tag":71,"props":23470,"children":23471},{},[23472],{"type":26,"value":23473},"IBMハードウェアのみをターゲットにしている",{"type":26,"value":23475}," — QiskitのトランスパイラとIBM Runtimeプリミティブは、IBMのQPU上で最高のパフォーマンスをもたらします。HLQuantumの抽象化はオーバーヘッドを追加します。",{"type":21,"tag":3560,"props":23477,"children":23478},{},[23479,23484],{"type":21,"tag":71,"props":23480,"children":23481},{},[23482],{"type":26,"value":23483},"きめ細かい回路制御が必要",{"type":26,"value":23485}," — カスタムなゲート分解、特定のトランスパイラパス、パルスレベルの制御、途中測定を伴う動的回路など。",{"type":21,"tag":3560,"props":23487,"children":23488},{},[23489,23494,23496,23502,23503,23509,23510,23516,23517,23523],{"type":21,"tag":71,"props":23490,"children":23491},{},[23492],{"type":26,"value":23493},"IBMエコシステムでの研究を行っている",{"type":26,"value":23495}," — ",{"type":21,"tag":79,"props":23497,"children":23499},{"className":23498},[],[23500],{"type":26,"value":23501},"qiskit-nature",{"type":26,"value":18579},{"type":21,"tag":79,"props":23504,"children":23506},{"className":23505},[],[23507],{"type":26,"value":23508},"qiskit-finance",{"type":26,"value":18579},{"type":21,"tag":79,"props":23511,"children":23513},{"className":23512},[],[23514],{"type":26,"value":23515},"qiskit-optimization",{"type":26,"value":18579},{"type":21,"tag":79,"props":23518,"children":23520},{"className":23519},[],[23521],{"type":26,"value":23522},"qiskit-machine-learning",{"type":26,"value":23524}," を使う場合。これらは素のQiskitと最も深く統合されています。",{"type":21,"tag":3560,"props":23526,"children":23527},{},[23528,23533],{"type":21,"tag":71,"props":23529,"children":23530},{},[23531],{"type":26,"value":23532},"パフォーマンスが重要",{"type":26,"value":23534}," — 直接のQiskitコードは、HLQuantumの変換レイヤーよりオーバーヘッドが少なくなります。",{"type":21,"tag":3560,"props":23536,"children":23537},{},[23538,23543],{"type":21,"tag":71,"props":23539,"children":23540},{},[23541],{"type":26,"value":23542},"Qiskitエコシステムに貢献している",{"type":26,"value":23544}," — Qiskitが参照実装となるプラグイン、チュートリアル、研究論文を書く場合。",{"type":21,"tag":59,"props":23546,"children":23548},{"id":23547},"hlquantumを使うべきとき",[23549],{"type":26,"value":23550},"HLQuantumを使うべきとき",{"type":21,"tag":22,"props":23552,"children":23553},{},[23554],{"type":26,"value":23555},"HLQuantumを使うべきなのは、次のような場合です。",{"type":21,"tag":7444,"props":23557,"children":23558},{},[23559,23569,23579,23589,23620],{"type":21,"tag":3560,"props":23560,"children":23561},{},[23562,23567],{"type":21,"tag":71,"props":23563,"children":23564},{},[23565],{"type":26,"value":23566},"バックエンドの柔軟性が必要",{"type":26,"value":23568}," — シミュレータの性能比較、異なるアーキテクチャでの回路ベンチマーク、あるいは複数のクラウドQPUへのアクセスなど。",{"type":21,"tag":3560,"props":23570,"children":23571},{},[23572,23577],{"type":21,"tag":71,"props":23573,"children":23574},{},[23575],{"type":26,"value":23576},"アルゴリズムのプロトタイプを作っている",{"type":26,"value":23578}," — 一度書けば、どこでもテストできます。ローカルシミュレーションからGPU、実機へと切り替えるときに回路を組み直す必要がありません。",{"type":21,"tag":3560,"props":23580,"children":23581},{},[23582,23587],{"type":21,"tag":71,"props":23583,"children":23584},{},[23585],{"type":26,"value":23586},"教えている、または学んでいる",{"type":26,"value":23588}," — すべてのバックエンドで一貫した1つのAPIが学習を容易にします。6種類の異なるAPIを学ぶ必要はありません。",{"type":21,"tag":3560,"props":23590,"children":23591},{},[23592,23597,23598,23604,23605,23611,23612,23618],{"type":21,"tag":71,"props":23593,"children":23594},{},[23595],{"type":26,"value":23596},"組み込みアルゴリズムを使いたい",{"type":26,"value":23495},{"type":21,"tag":79,"props":23599,"children":23601},{"className":23600},[],[23602],{"type":26,"value":23603},"hlq.algorithms.vqe()",{"type":26,"value":18579},{"type":21,"tag":79,"props":23606,"children":23608},{"className":23607},[],[23609],{"type":26,"value":23610},"hlq.algorithms.qaoa()",{"type":26,"value":18579},{"type":21,"tag":79,"props":23613,"children":23615},{"className":23614},[],[23616],{"type":26,"value":23617},"hlq.algorithms.grover()",{"type":26,"value":23619}," が定型処理を引き受けてくれます。",{"type":21,"tag":3560,"props":23621,"children":23622},{},[23623,23628],{"type":21,"tag":71,"props":23624,"children":23625},{},[23626],{"type":26,"value":23627},"マルチバックエンドのアプリケーションを構築している",{"type":26,"value":23629}," — アプリが、待ち行列が最も短い、あるいはコストが最も低いQPU上で実行される必要がある場合、HLQuantumのディスパッチャがそれを処理します。",{"type":21,"tag":59,"props":23631,"children":23633},{"id":23632},"移行-qiskit-hlquantum",[23634],{"type":26,"value":23635},"移行: Qiskit → HLQuantum",{"type":21,"tag":22,"props":23637,"children":23638},{},[23639],{"type":26,"value":23640},"既存のQiskitコードがある場合、HLQuantumはQiskit回路を直接取り込むことができます。",{"type":21,"tag":312,"props":23642,"children":23644},{"code":23643,"language":3628,"meta":7,"className":3626,"style":7},"import hlquantum as hlq\nfrom qiskit import QuantumCircuit\n\n# 既存のQiskit回路\nqc = QuantumCircuit(3)\nqc.h(0); qc.cx(0, 1); qc.cx(1, 2)\n\n# ラップして任意のバックエンドで実行\nresult = hlq.run(hlq.from_qiskit(qc), backend='pennylane')\nresult_gpu = hlq.run(hlq.from_qiskit(qc), backend='cudaq')\n",[23645],{"type":21,"tag":79,"props":23646,"children":23647},{"__ignoreMap":7},[23648,23667,23686,23693,23701,23724,23772,23779,23787,23819],{"type":21,"tag":322,"props":23649,"children":23650},{"class":324,"line":325},[23651,23655,23659,23663],{"type":21,"tag":322,"props":23652,"children":23653},{"style":3638},[23654],{"type":26,"value":3651},{"type":21,"tag":322,"props":23656,"children":23657},{"style":329},[23658],{"type":26,"value":19732},{"type":21,"tag":322,"props":23660,"children":23661},{"style":3638},[23662],{"type":26,"value":4715},{"type":21,"tag":322,"props":23664,"children":23665},{"style":329},[23666],{"type":26,"value":19741},{"type":21,"tag":322,"props":23668,"children":23669},{"class":324,"line":335},[23670,23674,23678,23682],{"type":21,"tag":322,"props":23671,"children":23672},{"style":3638},[23673],{"type":26,"value":3641},{"type":21,"tag":322,"props":23675,"children":23676},{"style":329},[23677],{"type":26,"value":4732},{"type":21,"tag":322,"props":23679,"children":23680},{"style":3638},[23681],{"type":26,"value":3651},{"type":21,"tag":322,"props":23683,"children":23684},{"style":329},[23685],{"type":26,"value":20739},{"type":21,"tag":322,"props":23687,"children":23688},{"class":324,"line":344},[23689],{"type":21,"tag":322,"props":23690,"children":23691},{"emptyLinePlaceholder":3662},[23692],{"type":26,"value":3665},{"type":21,"tag":322,"props":23694,"children":23695},{"class":324,"line":407},[23696],{"type":21,"tag":322,"props":23697,"children":23698},{"style":3888},[23699],{"type":26,"value":23700},"# 既存のQiskit回路\n",{"type":21,"tag":322,"props":23702,"children":23703},{"class":324,"line":431},[23704,23708,23712,23716,23720],{"type":21,"tag":322,"props":23705,"children":23706},{"style":329},[23707],{"type":26,"value":7819},{"type":21,"tag":322,"props":23709,"children":23710},{"style":3638},[23711],{"type":26,"value":3678},{"type":21,"tag":322,"props":23713,"children":23714},{"style":329},[23715],{"type":26,"value":5575},{"type":21,"tag":322,"props":23717,"children":23718},{"style":348},[23719],{"type":26,"value":6948},{"type":21,"tag":322,"props":23721,"children":23722},{"style":329},[23723],{"type":26,"value":3738},{"type":21,"tag":322,"props":23725,"children":23726},{"class":324,"line":453},[23727,23731,23735,23740,23744,23748,23752,23756,23760,23764,23768],{"type":21,"tag":322,"props":23728,"children":23729},{"style":329},[23730],{"type":26,"value":7835},{"type":21,"tag":322,"props":23732,"children":23733},{"style":348},[23734],{"type":26,"value":5605},{"type":21,"tag":322,"props":23736,"children":23737},{"style":329},[23738],{"type":26,"value":23739},"); qc.cx(",{"type":21,"tag":322,"props":23741,"children":23742},{"style":348},[23743],{"type":26,"value":5605},{"type":21,"tag":322,"props":23745,"children":23746},{"style":329},[23747],{"type":26,"value":367},{"type":21,"tag":322,"props":23749,"children":23750},{"style":348},[23751],{"type":26,"value":3994},{"type":21,"tag":322,"props":23753,"children":23754},{"style":329},[23755],{"type":26,"value":23739},{"type":21,"tag":322,"props":23757,"children":23758},{"style":348},[23759],{"type":26,"value":3994},{"type":21,"tag":322,"props":23761,"children":23762},{"style":329},[23763],{"type":26,"value":367},{"type":21,"tag":322,"props":23765,"children":23766},{"style":348},[23767],{"type":26,"value":423},{"type":21,"tag":322,"props":23769,"children":23770},{"style":329},[23771],{"type":26,"value":3738},{"type":21,"tag":322,"props":23773,"children":23774},{"class":324,"line":474},[23775],{"type":21,"tag":322,"props":23776,"children":23777},{"emptyLinePlaceholder":3662},[23778],{"type":26,"value":3665},{"type":21,"tag":322,"props":23780,"children":23781},{"class":324,"line":496},[23782],{"type":21,"tag":322,"props":23783,"children":23784},{"style":3888},[23785],{"type":26,"value":23786},"# ラップして任意のバックエンドで実行\n",{"type":21,"tag":322,"props":23788,"children":23789},{"class":324,"line":16},[23790,23794,23798,23803,23807,23811,23815],{"type":21,"tag":322,"props":23791,"children":23792},{"style":329},[23793],{"type":26,"value":19764},{"type":21,"tag":322,"props":23795,"children":23796},{"style":3638},[23797],{"type":26,"value":3678},{"type":21,"tag":322,"props":23799,"children":23800},{"style":329},[23801],{"type":26,"value":23802}," hlq.run(hlq.from_qiskit(qc), ",{"type":21,"tag":322,"props":23804,"children":23805},{"style":3703},[23806],{"type":26,"value":18908},{"type":21,"tag":322,"props":23808,"children":23809},{"style":3638},[23810],{"type":26,"value":3678},{"type":21,"tag":322,"props":23812,"children":23813},{"style":359},[23814],{"type":26,"value":22610},{"type":21,"tag":322,"props":23816,"children":23817},{"style":329},[23818],{"type":26,"value":3738},{"type":21,"tag":322,"props":23820,"children":23821},{"class":324,"line":3884},[23822,23826,23830,23834,23838,23842,23846],{"type":21,"tag":322,"props":23823,"children":23824},{"style":329},[23825],{"type":26,"value":23406},{"type":21,"tag":322,"props":23827,"children":23828},{"style":3638},[23829],{"type":26,"value":3678},{"type":21,"tag":322,"props":23831,"children":23832},{"style":329},[23833],{"type":26,"value":23802},{"type":21,"tag":322,"props":23835,"children":23836},{"style":3703},[23837],{"type":26,"value":18908},{"type":21,"tag":322,"props":23839,"children":23840},{"style":3638},[23841],{"type":26,"value":3678},{"type":21,"tag":322,"props":23843,"children":23844},{"style":359},[23845],{"type":26,"value":22643},{"type":21,"tag":322,"props":23847,"children":23848},{"style":329},[23849],{"type":26,"value":3738},{"type":21,"tag":22,"props":23851,"children":23852},{},[23853],{"type":26,"value":23854},"書き直しは不要です。既存のQiskitコードに、HLQuantumのマルチバックエンド機能を段階的に追加できます。",{"type":21,"tag":59,"props":23856,"children":23858},{"id":23857},"判断チェックリスト",[23859],{"type":26,"value":23857},{"type":21,"tag":312,"props":23861,"children":23863},{"code":23862},"Starting a new project?\n├── Need IBM-specific features (Qiskit Runtime, pulse control)?  → Raw Qiskit\n├── Need to run on multiple backends or compare performance?     → HLQuantum\n├── Learning quantum computing for the first time?               → HLQuantum\n└── Building a research paper targeting IBM hardware?            → Raw Qiskit\n\nMigrating existing Qiskit code?\n├── Want to add GPU acceleration?     → HLQuantum (from_qiskit + cudaq)\n├── Want to try IonQ hardware?        → HLQuantum (from_qiskit + ionq)\n└── Want to stay IBM-only?            → Stay with Qiskit\n",[23864],{"type":21,"tag":79,"props":23865,"children":23866},{"__ignoreMap":7},[23867],{"type":26,"value":23862},{"type":21,"tag":22,"props":23869,"children":23870},{},[23871],{"type":26,"value":23872},"両者は互いに排他的なものではありません。多くのワークフローは、回路設計には素のQiskitで始め、その後マルチバックエンドのデプロイのためにHLQuantumでラップします。",{"type":21,"tag":22,"props":23874,"children":23875},{},[23876,23881,23882,23887,23888,23892,23893],{"type":21,"tag":71,"props":23877,"children":23878},{},[23879],{"type":26,"value":23880},"始めましょう:",{"type":26,"value":1901},{"type":21,"tag":29,"props":23883,"children":23884},{"href":19441},[23885],{"type":26,"value":23886},"HLQuantum guide",{"type":26,"value":7549},{"type":21,"tag":29,"props":23889,"children":23890},{"href":7528},[23891],{"type":26,"value":9240},{"type":26,"value":7549},{"type":21,"tag":29,"props":23894,"children":23895},{"href":814},[23896],{"type":26,"value":23897},"Compare all SDKs",{"type":21,"tag":703,"props":23899,"children":23900},{},[23901],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":23903},[23904,23905,23906,23907,23908,23909,23910,23911],{"id":21470,"depth":335,"text":21470},{"id":21493,"depth":335,"text":21496},{"id":21913,"depth":335,"text":21916},{"id":22793,"depth":335,"text":22796},{"id":23454,"depth":335,"text":23457},{"id":23547,"depth":335,"text":23550},{"id":23632,"depth":335,"text":23635},{"id":23857,"depth":335,"text":23857},"content:ja:blog:hlquantum-vs-qiskit.md","ja\u002Fblog\u002Fhlquantum-vs-qiskit.md","ja\u002Fblog\u002Fhlquantum-vs-qiskit",{"_path":23916,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":23917,"description":23918,"date":23919,"author":11,"tags":23920,"readingTime":3884,"body":23921,"_type":715,"_id":24964,"_source":717,"_file":24965,"_stem":24966,"_extension":720},"\u002Fja\u002Fblog\u002Fpost-quantum-cryptography-guide","ポスト量子暗号: 開発者が今すぐ取り組むべきこと","Shor のアルゴリズムは RSA と ECC を脅かします。NIST はポスト量子標準を最終化しました。ここでは、その脅威が何であるか、標準が何を規定しているか、そして開発者が今日どのような行動を取るべきかを解説します。","2026-03-05",[9786,9787,9785],{"type":18,"children":23922,"toc":24948},[23923,23935,23940,23946,23959,23967,23990,23998,24021,24033,24039,24044,24062,24078,24086,24114,24122,24140,24146,24151,24157,24200,24361,24367,24406,24529,24535,24574,24586,24592,24598,24608,24631,24641,24651,24657,24675,24700,24710,24720,24726,24736,24746,24756,24761,24905,24911,24919,24924,24944],{"type":21,"tag":22,"props":23924,"children":23925},{},[23926,23928,23933],{"type":26,"value":23927},"量子コンピュータはまだあなたの暗号を脅かしてはいません。RSA-2048 を破れる量子コンピュータには、誤り訂正された論理量子ビットが数百万個必要ですが、現在私たちが手にしているのは、ノイズの多い物理量子ビットが数百個です。しかし、ここに問題があります。敵対者は、量子ハードウェアが成熟した後に復号することを狙って、",{"type":21,"tag":71,"props":23929,"children":23930},{},[23931],{"type":26,"value":23932},"今この瞬間",{"type":26,"value":23934},"に暗号化されたデータを収集している可能性があるのです。そして、暗号インフラの移行には何年もかかります。",{"type":21,"tag":22,"props":23936,"children":23937},{},[23938],{"type":26,"value":23939},"これこそが、2024 年の NIST ポスト量子暗号標準が、2035 年ではなく今日の開発者にとって重要である理由です。",{"type":21,"tag":59,"props":23941,"children":23943},{"id":23942},"実際の脅威-shor-のアルゴリズム",[23944],{"type":26,"value":23945},"実際の脅威: Shor のアルゴリズム",{"type":21,"tag":22,"props":23947,"children":23948},{},[23949,23951,23957],{"type":26,"value":23950},"1994 年、Peter Shor は、量子コンピュータが大きな整数を多項式時間 ",{"type":21,"tag":79,"props":23952,"children":23954},{"className":23953},[],[23955],{"type":26,"value":23956},"O((log N)³)",{"type":26,"value":23958}," で素因数分解できることを証明しました。これは、素因数分解が困難であることに安全性を依存している RSA を破ります。同じアルゴリズムは、わずかな変更を加えることで、ECDSA と ECDH の基盤である楕円曲線暗号 (ECC) をも破ります。",{"type":21,"tag":22,"props":23960,"children":23961},{},[23962],{"type":21,"tag":71,"props":23963,"children":23964},{},[23965],{"type":26,"value":23966},"破られるもの:",{"type":21,"tag":3556,"props":23968,"children":23969},{},[23970,23975,23980,23985],{"type":21,"tag":3560,"props":23971,"children":23972},{},[23973],{"type":26,"value":23974},"RSA (暗号化、署名)",{"type":21,"tag":3560,"props":23976,"children":23977},{},[23978],{"type":26,"value":23979},"ECDSA (署名 — TLS、SSH、コード署名、Bitcoin で使用)",{"type":21,"tag":3560,"props":23981,"children":23982},{},[23983],{"type":26,"value":23984},"ECDH (鍵交換 — TLS、Signal、WhatsApp、iMessage で使用)",{"type":21,"tag":3560,"props":23986,"children":23987},{},[23988],{"type":26,"value":23989},"Diffie-Hellman (鍵交換)",{"type":21,"tag":22,"props":23991,"children":23992},{},[23993],{"type":21,"tag":71,"props":23994,"children":23995},{},[23996],{"type":26,"value":23997},"破られないもの:",{"type":21,"tag":3556,"props":23999,"children":24000},{},[24001,24006,24011,24016],{"type":21,"tag":3560,"props":24002,"children":24003},{},[24004],{"type":26,"value":24005},"AES-256 (共通鍵暗号) — Grover のアルゴリズムは二次的な高速化しか提供せず、実効的な鍵強度を 256 ビットから 128 ビットに減らすだけです。これは対処可能です。",{"type":21,"tag":3560,"props":24007,"children":24008},{},[24009],{"type":26,"value":24010},"SHA-256、SHA-3 (ハッシュ関数) — 同様に、Grover はわずかな高速化を提供し、安全性は半減しますが、現在の鍵サイズであれば依然として十分に強固です。",{"type":21,"tag":3560,"props":24012,"children":24013},{},[24014],{"type":26,"value":24015},"HMAC — 安全",{"type":21,"tag":3560,"props":24017,"children":24018},{},[24019],{"type":26,"value":24020},"Argon2、bcrypt (パスワードハッシュ化) — 安全",{"type":21,"tag":22,"props":24022,"children":24023},{},[24024,24026,24031],{"type":26,"value":24025},"結論はこうです。",{"type":21,"tag":71,"props":24027,"children":24028},{},[24029],{"type":26,"value":24030},"非対称暗号は置き換えが必要であり、対称暗号は鍵サイズの引き上げ (AES-128 → AES-256) が必要です",{"type":26,"value":24032},"。",{"type":21,"tag":59,"props":24034,"children":24036},{"id":24035},"今収穫し後で復号する-harvest-now-decrypt-later",[24037],{"type":26,"value":24038},"今収穫し、後で復号する (Harvest Now, Decrypt Later)",{"type":21,"tag":22,"props":24040,"children":24041},{},[24042],{"type":26,"value":24043},"脅威は「量子コンピュータが来年には暗号を破る」ということではありません。脅威とは次のようなものです。",{"type":21,"tag":7444,"props":24045,"children":24046},{},[24047,24052,24057],{"type":21,"tag":3560,"props":24048,"children":24049},{},[24050],{"type":26,"value":24051},"敵対者が今日、あなたの暗号化された TLS トラフィックを捕捉して保存する",{"type":21,"tag":3560,"props":24053,"children":24054},{},[24055],{"type":26,"value":24056},"2032〜2040 年に、それに対して Shor のアルゴリズムを実行する",{"type":21,"tag":3560,"props":24058,"children":24059},{},[24060],{"type":26,"value":24061},"すべてを遡って復号する",{"type":21,"tag":22,"props":24063,"children":24064},{},[24065,24070,24071,24076],{"type":21,"tag":71,"props":24066,"children":24067},{},[24068],{"type":26,"value":24069},"「今収穫し、後で復号する」",{"type":26,"value":19246},{"type":21,"tag":71,"props":24072,"children":24073},{},[24074],{"type":26,"value":24075},"HNDL",{"type":26,"value":24077}," と呼ばれるこの攻撃は、すでに起きています。NSA やその他の諜報機関は、ほぼ間違いなく大量の暗号化トラフィックを保存しています。2022 年の NSA の勧告は、これについて明確に警告しました。",{"type":21,"tag":22,"props":24079,"children":24080},{},[24081],{"type":21,"tag":71,"props":24082,"children":24083},{},[24084],{"type":26,"value":24085},"今日 HNDL に対して保護が必要なデータ:",{"type":21,"tag":3556,"props":24087,"children":24088},{},[24089,24094,24099,24104,24109],{"type":21,"tag":3560,"props":24090,"children":24091},{},[24092],{"type":26,"value":24093},"国家機密、機密扱いの政府データ",{"type":21,"tag":3560,"props":24095,"children":24096},{},[24097],{"type":26,"value":24098},"長期にわたって有効な個人データ (医療記録、身分証明書類)",{"type":21,"tag":3560,"props":24100,"children":24101},{},[24102],{"type":26,"value":24103},"長期の法的保存要件を伴う金融データ",{"type":21,"tag":3560,"props":24105,"children":24106},{},[24107],{"type":26,"value":24108},"営業秘密と知的財産",{"type":21,"tag":3560,"props":24110,"children":24111},{},[24112],{"type":26,"value":24113},"パスワードデータベース (ただし bcrypt\u002FArgon2 は安全)",{"type":21,"tag":22,"props":24115,"children":24116},{},[24117],{"type":21,"tag":71,"props":24118,"children":24119},{},[24120],{"type":26,"value":24121},"すぐに心配する必要のないデータ:",{"type":21,"tag":3556,"props":24123,"children":24124},{},[24125,24130,24135],{"type":21,"tag":3560,"props":24126,"children":24127},{},[24128],{"type":26,"value":24129},"セッショントークン (短命で、10 年後には無価値)",{"type":21,"tag":3560,"props":24131,"children":24132},{},[24133],{"type":26,"value":24134},"一時的なチャットメッセージ (あなたが高価値の標的でない限り)",{"type":21,"tag":3560,"props":24136,"children":24137},{},[24138],{"type":26,"value":24139},"公開情報",{"type":21,"tag":59,"props":24141,"children":24143},{"id":24142},"nist-のポスト量子標準-2024",[24144],{"type":26,"value":24145},"NIST のポスト量子標準 (2024)",{"type":21,"tag":22,"props":24147,"children":24148},{},[24149],{"type":26,"value":24150},"8 年にわたるコンペティションを経て、NIST は 2024 年 8 月に 3 つのポスト量子暗号アルゴリズムを最終化しました。",{"type":21,"tag":17800,"props":24152,"children":24154},{"id":24153},"ml-kem-crystals-kyber-鍵カプセル化",[24155],{"type":26,"value":24156},"ML-KEM (CRYSTALS-Kyber) — 鍵カプセル化",{"type":21,"tag":3556,"props":24158,"children":24159},{},[24160,24170,24180,24190],{"type":21,"tag":3560,"props":24161,"children":24162},{},[24163,24168],{"type":21,"tag":71,"props":24164,"children":24165},{},[24166],{"type":26,"value":24167},"置き換え対象:",{"type":26,"value":24169}," RSA 鍵交換、ECDH",{"type":21,"tag":3560,"props":24171,"children":24172},{},[24173,24178],{"type":21,"tag":71,"props":24174,"children":24175},{},[24176],{"type":26,"value":24177},"安全性の根拠:",{"type":26,"value":24179}," Module Learning With Errors (MLWE) 問題 — 古典コンピュータと量子コンピュータの双方にとって困難であると考えられている",{"type":21,"tag":3560,"props":24181,"children":24182},{},[24183,24188],{"type":21,"tag":71,"props":24184,"children":24185},{},[24186],{"type":26,"value":24187},"性能:",{"type":26,"value":24189}," 高速、小さな鍵 (KEM-768 で約 800 バイト)、高速な演算",{"type":21,"tag":3560,"props":24191,"children":24192},{},[24193,24198],{"type":21,"tag":71,"props":24194,"children":24195},{},[24196],{"type":26,"value":24197},"用途:",{"type":26,"value":24199}," TLS 鍵交換、暗号化された鍵ラッピング、ハイブリッド暗号方式",{"type":21,"tag":312,"props":24201,"children":24203},{"code":24202,"language":3628,"meta":7,"className":3626,"style":7},"# Python example using pqcrypto library\nfrom pqcrypto.kem.kyber768 import generate_keypair, encapsulate, decapsulate\n\npublic_key, secret_key = generate_keypair()\n\n# Sender encapsulates a shared secret\nciphertext, shared_secret_sender = encapsulate(public_key)\n\n# Receiver decapsulates\nshared_secret_receiver = decapsulate(secret_key, ciphertext)\n\nassert shared_secret_sender == shared_secret_receiver\n# Now use shared_secret as AES key\n",[24204],{"type":21,"tag":79,"props":24205,"children":24206},{"__ignoreMap":7},[24207,24215,24236,24243,24260,24267,24275,24292,24299,24307,24324,24331,24353],{"type":21,"tag":322,"props":24208,"children":24209},{"class":324,"line":325},[24210],{"type":21,"tag":322,"props":24211,"children":24212},{"style":3888},[24213],{"type":26,"value":24214},"# Python example using pqcrypto library\n",{"type":21,"tag":322,"props":24216,"children":24217},{"class":324,"line":335},[24218,24222,24227,24231],{"type":21,"tag":322,"props":24219,"children":24220},{"style":3638},[24221],{"type":26,"value":3641},{"type":21,"tag":322,"props":24223,"children":24224},{"style":329},[24225],{"type":26,"value":24226}," pqcrypto.kem.kyber768 ",{"type":21,"tag":322,"props":24228,"children":24229},{"style":3638},[24230],{"type":26,"value":3651},{"type":21,"tag":322,"props":24232,"children":24233},{"style":329},[24234],{"type":26,"value":24235}," generate_keypair, encapsulate, decapsulate\n",{"type":21,"tag":322,"props":24237,"children":24238},{"class":324,"line":344},[24239],{"type":21,"tag":322,"props":24240,"children":24241},{"emptyLinePlaceholder":3662},[24242],{"type":26,"value":3665},{"type":21,"tag":322,"props":24244,"children":24245},{"class":324,"line":407},[24246,24251,24255],{"type":21,"tag":322,"props":24247,"children":24248},{"style":329},[24249],{"type":26,"value":24250},"public_key, secret_key ",{"type":21,"tag":322,"props":24252,"children":24253},{"style":3638},[24254],{"type":26,"value":3678},{"type":21,"tag":322,"props":24256,"children":24257},{"style":329},[24258],{"type":26,"value":24259}," generate_keypair()\n",{"type":21,"tag":322,"props":24261,"children":24262},{"class":324,"line":431},[24263],{"type":21,"tag":322,"props":24264,"children":24265},{"emptyLinePlaceholder":3662},[24266],{"type":26,"value":3665},{"type":21,"tag":322,"props":24268,"children":24269},{"class":324,"line":453},[24270],{"type":21,"tag":322,"props":24271,"children":24272},{"style":3888},[24273],{"type":26,"value":24274},"# Sender encapsulates a shared secret\n",{"type":21,"tag":322,"props":24276,"children":24277},{"class":324,"line":474},[24278,24283,24287],{"type":21,"tag":322,"props":24279,"children":24280},{"style":329},[24281],{"type":26,"value":24282},"ciphertext, shared_secret_sender ",{"type":21,"tag":322,"props":24284,"children":24285},{"style":3638},[24286],{"type":26,"value":3678},{"type":21,"tag":322,"props":24288,"children":24289},{"style":329},[24290],{"type":26,"value":24291}," encapsulate(public_key)\n",{"type":21,"tag":322,"props":24293,"children":24294},{"class":324,"line":496},[24295],{"type":21,"tag":322,"props":24296,"children":24297},{"emptyLinePlaceholder":3662},[24298],{"type":26,"value":3665},{"type":21,"tag":322,"props":24300,"children":24301},{"class":324,"line":16},[24302],{"type":21,"tag":322,"props":24303,"children":24304},{"style":3888},[24305],{"type":26,"value":24306},"# Receiver decapsulates\n",{"type":21,"tag":322,"props":24308,"children":24309},{"class":324,"line":3884},[24310,24315,24319],{"type":21,"tag":322,"props":24311,"children":24312},{"style":329},[24313],{"type":26,"value":24314},"shared_secret_receiver ",{"type":21,"tag":322,"props":24316,"children":24317},{"style":3638},[24318],{"type":26,"value":3678},{"type":21,"tag":322,"props":24320,"children":24321},{"style":329},[24322],{"type":26,"value":24323}," decapsulate(secret_key, ciphertext)\n",{"type":21,"tag":322,"props":24325,"children":24326},{"class":324,"line":3894},[24327],{"type":21,"tag":322,"props":24328,"children":24329},{"emptyLinePlaceholder":3662},[24330],{"type":26,"value":3665},{"type":21,"tag":322,"props":24332,"children":24333},{"class":324,"line":1034},[24334,24339,24344,24348],{"type":21,"tag":322,"props":24335,"children":24336},{"style":3638},[24337],{"type":26,"value":24338},"assert",{"type":21,"tag":322,"props":24340,"children":24341},{"style":329},[24342],{"type":26,"value":24343}," shared_secret_sender ",{"type":21,"tag":322,"props":24345,"children":24346},{"style":3638},[24347],{"type":26,"value":5346},{"type":21,"tag":322,"props":24349,"children":24350},{"style":329},[24351],{"type":26,"value":24352}," shared_secret_receiver\n",{"type":21,"tag":322,"props":24354,"children":24355},{"class":324,"line":3954},[24356],{"type":21,"tag":322,"props":24357,"children":24358},{"style":3888},[24359],{"type":26,"value":24360},"# Now use shared_secret as AES key\n",{"type":21,"tag":17800,"props":24362,"children":24364},{"id":24363},"ml-dsa-crystals-dilithium-デジタル署名",[24365],{"type":26,"value":24366},"ML-DSA (CRYSTALS-Dilithium) — デジタル署名",{"type":21,"tag":3556,"props":24368,"children":24369},{},[24370,24379,24388,24397],{"type":21,"tag":3560,"props":24371,"children":24372},{},[24373,24377],{"type":21,"tag":71,"props":24374,"children":24375},{},[24376],{"type":26,"value":24167},{"type":26,"value":24378}," 署名のための ECDSA、RSA-PSS",{"type":21,"tag":3560,"props":24380,"children":24381},{},[24382,24386],{"type":21,"tag":71,"props":24383,"children":24384},{},[24385],{"type":26,"value":24177},{"type":26,"value":24387}," MLWE + MSIS 問題",{"type":21,"tag":3560,"props":24389,"children":24390},{},[24391,24395],{"type":21,"tag":71,"props":24392,"children":24393},{},[24394],{"type":26,"value":24187},{"type":26,"value":24396}," 中程度の鍵サイズ (公開鍵で約 1,312 バイト)、高速な署名",{"type":21,"tag":3560,"props":24398,"children":24399},{},[24400,24404],{"type":21,"tag":71,"props":24401,"children":24402},{},[24403],{"type":26,"value":24197},{"type":26,"value":24405}," コード署名、TLS 証明書、文書署名、JWT 署名",{"type":21,"tag":312,"props":24407,"children":24409},{"code":24408,"language":3628,"meta":7,"className":3626,"style":7},"from pqcrypto.sign.dilithium3 import generate_keypair, sign, verify\n\npublic_key, secret_key = generate_keypair()\n\nmessage = b\"Deploy to production\"\nsignature = sign(message, secret_key)\n\n# Verify\nassert verify(message, signature, public_key)\n",[24410],{"type":21,"tag":79,"props":24411,"children":24412},{"__ignoreMap":7},[24413,24434,24441,24456,24463,24485,24502,24509,24517],{"type":21,"tag":322,"props":24414,"children":24415},{"class":324,"line":325},[24416,24420,24425,24429],{"type":21,"tag":322,"props":24417,"children":24418},{"style":3638},[24419],{"type":26,"value":3641},{"type":21,"tag":322,"props":24421,"children":24422},{"style":329},[24423],{"type":26,"value":24424}," pqcrypto.sign.dilithium3 ",{"type":21,"tag":322,"props":24426,"children":24427},{"style":3638},[24428],{"type":26,"value":3651},{"type":21,"tag":322,"props":24430,"children":24431},{"style":329},[24432],{"type":26,"value":24433}," generate_keypair, sign, verify\n",{"type":21,"tag":322,"props":24435,"children":24436},{"class":324,"line":335},[24437],{"type":21,"tag":322,"props":24438,"children":24439},{"emptyLinePlaceholder":3662},[24440],{"type":26,"value":3665},{"type":21,"tag":322,"props":24442,"children":24443},{"class":324,"line":344},[24444,24448,24452],{"type":21,"tag":322,"props":24445,"children":24446},{"style":329},[24447],{"type":26,"value":24250},{"type":21,"tag":322,"props":24449,"children":24450},{"style":3638},[24451],{"type":26,"value":3678},{"type":21,"tag":322,"props":24453,"children":24454},{"style":329},[24455],{"type":26,"value":24259},{"type":21,"tag":322,"props":24457,"children":24458},{"class":324,"line":407},[24459],{"type":21,"tag":322,"props":24460,"children":24461},{"emptyLinePlaceholder":3662},[24462],{"type":26,"value":3665},{"type":21,"tag":322,"props":24464,"children":24465},{"class":324,"line":431},[24466,24471,24475,24480],{"type":21,"tag":322,"props":24467,"children":24468},{"style":329},[24469],{"type":26,"value":24470},"message ",{"type":21,"tag":322,"props":24472,"children":24473},{"style":3638},[24474],{"type":26,"value":3678},{"type":21,"tag":322,"props":24476,"children":24477},{"style":3638},[24478],{"type":26,"value":24479}," b",{"type":21,"tag":322,"props":24481,"children":24482},{"style":359},[24483],{"type":26,"value":24484},"\"Deploy to production\"\n",{"type":21,"tag":322,"props":24486,"children":24487},{"class":324,"line":453},[24488,24493,24497],{"type":21,"tag":322,"props":24489,"children":24490},{"style":329},[24491],{"type":26,"value":24492},"signature ",{"type":21,"tag":322,"props":24494,"children":24495},{"style":3638},[24496],{"type":26,"value":3678},{"type":21,"tag":322,"props":24498,"children":24499},{"style":329},[24500],{"type":26,"value":24501}," sign(message, secret_key)\n",{"type":21,"tag":322,"props":24503,"children":24504},{"class":324,"line":474},[24505],{"type":21,"tag":322,"props":24506,"children":24507},{"emptyLinePlaceholder":3662},[24508],{"type":26,"value":3665},{"type":21,"tag":322,"props":24510,"children":24511},{"class":324,"line":496},[24512],{"type":21,"tag":322,"props":24513,"children":24514},{"style":3888},[24515],{"type":26,"value":24516},"# Verify\n",{"type":21,"tag":322,"props":24518,"children":24519},{"class":324,"line":16},[24520,24524],{"type":21,"tag":322,"props":24521,"children":24522},{"style":3638},[24523],{"type":26,"value":24338},{"type":21,"tag":322,"props":24525,"children":24526},{"style":329},[24527],{"type":26,"value":24528}," verify(message, signature, public_key)\n",{"type":21,"tag":17800,"props":24530,"children":24532},{"id":24531},"slh-dsa-sphincs-ハッシュベースの署名",[24533],{"type":26,"value":24534},"SLH-DSA (SPHINCS+) — ハッシュベースの署名",{"type":21,"tag":3556,"props":24536,"children":24537},{},[24538,24547,24556,24565],{"type":21,"tag":3560,"props":24539,"children":24540},{},[24541,24545],{"type":21,"tag":71,"props":24542,"children":24543},{},[24544],{"type":26,"value":24167},{"type":26,"value":24546}," 保守的なバックアップとしての ECDSA",{"type":21,"tag":3560,"props":24548,"children":24549},{},[24550,24554],{"type":21,"tag":71,"props":24551,"children":24552},{},[24553],{"type":26,"value":24177},{"type":26,"value":24555}," ハッシュ関数のみ — 最も保守的な選択肢",{"type":21,"tag":3560,"props":24557,"children":24558},{},[24559,24563],{"type":21,"tag":71,"props":24560,"children":24561},{},[24562],{"type":26,"value":24187},{"type":26,"value":24564}," より大きな署名 (約 8〜50 KB)、より遅い署名処理",{"type":21,"tag":3560,"props":24566,"children":24567},{},[24568,24572],{"type":21,"tag":71,"props":24569,"children":24570},{},[24571],{"type":26,"value":24197},{"type":26,"value":24573}," 鍵の再利用が懸念される高保証の署名、長期の文書署名",{"type":21,"tag":22,"props":24575,"children":24576},{},[24577,24579,24584],{"type":26,"value":24578},"NIST はまた、",{"type":21,"tag":71,"props":24580,"children":24581},{},[24582],{"type":26,"value":24583},"FALCON (FN-DSA)",{"type":26,"value":24585}," も標準化しました。これはコンパクトな署名を持つ格子ベースの署名方式で、IoT のような制約の多い環境に適しています。",{"type":21,"tag":59,"props":24587,"children":24589},{"id":24588},"今すぐやるべきこと-段階別のアクションプラン",[24590],{"type":26,"value":24591},"今すぐやるべきこと: 段階別のアクションプラン",{"type":21,"tag":17800,"props":24593,"children":24595},{"id":24594},"第-1-段階-今日やること-低労力高インパクト",[24596],{"type":26,"value":24597},"第 1 段階 — 今日やること (低労力・高インパクト)",{"type":21,"tag":22,"props":24599,"children":24600},{},[24601,24606],{"type":21,"tag":71,"props":24602,"children":24603},{},[24604],{"type":26,"value":24605},"暗号インベントリを監査する。",{"type":26,"value":24607}," カタログ化していないものは移行できません。次を文書化しましょう。",{"type":21,"tag":3556,"props":24609,"children":24610},{},[24611,24616,24621,24626],{"type":21,"tag":3560,"props":24612,"children":24613},{},[24614],{"type":26,"value":24615},"アプリケーションはどこで TLS を使用しているか? (ほとんどのフレームワークがこれを扱う)",{"type":21,"tag":3560,"props":24617,"children":24618},{},[24619],{"type":26,"value":24620},"どこで署名を生成または検証しているか? (JWT、コード署名、PDF 署名)",{"type":21,"tag":3560,"props":24622,"children":24623},{},[24624],{"type":26,"value":24625},"どこで非対称鍵交換を使用しているか? (RSA でラップされた AES 鍵、SSH)",{"type":21,"tag":3560,"props":24627,"children":24628},{},[24629],{"type":26,"value":24630},"どのライブラリを使用しているか? (OpenSSL、BouncyCastle、AWS KMS など)",{"type":21,"tag":22,"props":24632,"children":24633},{},[24634,24639],{"type":21,"tag":71,"props":24635,"children":24636},{},[24637],{"type":26,"value":24638},"AES-128 を AES-256 にアップグレードする。",{"type":26,"value":24640}," これは低リスクで即座に実施できます。AES-256 は Grover に対して 128 ビットの安全性に半減しますが、それでも十分に安全です。",{"type":21,"tag":22,"props":24642,"children":24643},{},[24644,24649],{"type":21,"tag":71,"props":24645,"children":24646},{},[24647],{"type":26,"value":24648},"あらゆる場所で TLS 1.3 を有効にする。",{"type":26,"value":24650}," TLS 1.3 はデフォルトで前方秘匿性 (ECDHE) を使用します。これは、長期鍵が漏洩したとしても過去のセッションを復号できないことを意味します。これにより、TLS トラフィックに対する HNDL が部分的に緩和されます。",{"type":21,"tag":17800,"props":24652,"children":24654},{"id":24653},"第-2-段階-今後-618-か月の計画",[24655],{"type":26,"value":24656},"第 2 段階 — 今後 6〜18 か月の計画",{"type":21,"tag":22,"props":24658,"children":24659},{},[24660,24665,24667,24673],{"type":21,"tag":71,"props":24661,"children":24662},{},[24663],{"type":26,"value":24664},"TLS でハイブリッド鍵交換を採用する。",{"type":26,"value":24666}," TLS ワーキンググループは ",{"type":21,"tag":79,"props":24668,"children":24670},{"className":24669},[],[24671],{"type":26,"value":24672},"X25519Kyber768",{"type":26,"value":24674}," を標準化しました。これは古典的な ECDH と ML-KEM の双方を使用するハイブリッド方式です。Google Chrome、Cloudflare、AWS はすでにこれをサポートしています。これは古典的な安全性を維持しつつ HNDL から保護します。",{"type":21,"tag":312,"props":24676,"children":24680},{"code":24677,"language":24678,"meta":7,"className":24679,"style":7},"# Nginx with OpenSSL 3.x + oqs-provider (Open Quantum Safe)\nssl_ecdh_curve X25519Kyber768;  # hybrid classical + PQC\n","nginx","language-nginx shiki shiki-themes github-dark",[24681],{"type":21,"tag":79,"props":24682,"children":24683},{"__ignoreMap":7},[24684,24692],{"type":21,"tag":322,"props":24685,"children":24686},{"class":324,"line":325},[24687],{"type":21,"tag":322,"props":24688,"children":24689},{},[24690],{"type":26,"value":24691},"# Nginx with OpenSSL 3.x + oqs-provider (Open Quantum Safe)\n",{"type":21,"tag":322,"props":24693,"children":24694},{"class":324,"line":335},[24695],{"type":21,"tag":322,"props":24696,"children":24697},{},[24698],{"type":26,"value":24699},"ssl_ecdh_curve X25519Kyber768;  # hybrid classical + PQC\n",{"type":21,"tag":22,"props":24701,"children":24702},{},[24703,24708],{"type":21,"tag":71,"props":24704,"children":24705},{},[24706],{"type":26,"value":24707},"新しいシステムでは RSA\u002FECDSA を ML-DSA に置き換える。",{"type":26,"value":24709}," 新しいマイクロサービス、新しい JWT 実装、新しい SSH デプロイなど、あらゆる新規構築のものには、ECDSA-256 の代わりに ML-DSA (Dilithium3 または Dilithium5) を使用しましょう。",{"type":21,"tag":22,"props":24711,"children":24712},{},[24713,24718],{"type":21,"tag":71,"props":24714,"children":24715},{},[24716],{"type":26,"value":24717},"TLS 証明書のパイプラインを更新する。",{"type":26,"value":24719}," ほとんどのパブリック CA はまだ PQC 証明書を発行していませんが、内部署名された証明書やサービスメッシュ証明書には、今日から ML-DSA を使用できます。",{"type":21,"tag":17800,"props":24721,"children":24723},{"id":24722},"第-3-段階-長期的な移行-15-年",[24724],{"type":26,"value":24725},"第 3 段階 — 長期的な移行 (1〜5 年)",{"type":21,"tag":22,"props":24727,"children":24728},{},[24729,24734],{"type":21,"tag":71,"props":24730,"children":24731},{},[24732],{"type":26,"value":24733},"完全な PKI 移行。",{"type":26,"value":24735}," RSA\u002FECC の証明書階層を PQC ベースの CA インフラに置き換えます。これは最も難しい部分です。ハードウェアセキュリティモジュール (HSM)、自動化された証明書ローテーション、証明書チェーンのすべてを更新する必要があります。",{"type":21,"tag":22,"props":24737,"children":24738},{},[24739,24744],{"type":21,"tag":71,"props":24740,"children":24741},{},[24742],{"type":26,"value":24743},"ハードウェアと組み込みデバイス。",{"type":26,"value":24745}," IoT や組み込みデバイスは、多くの場合リモートでパッチを当てられず、10 年以上の寿命を持ちます。RSA や ECDSA に認証を依存しているそのようなデバイスを監査しましょう。",{"type":21,"tag":22,"props":24747,"children":24748},{},[24749,24754],{"type":21,"tag":71,"props":24750,"children":24751},{},[24752],{"type":26,"value":24753},"鍵管理システム。",{"type":26,"value":24755}," AWS KMS、Azure Key Vault、HashiCorp Vault — それらの PQC ロードマップと移行スケジュールを確認しましょう。",{"type":21,"tag":59,"props":24757,"children":24759},{"id":24758},"ライブラリとフレームワークのサポート状況",[24760],{"type":26,"value":24758},{"type":21,"tag":4503,"props":24762,"children":24763},{},[24764,24780],{"type":21,"tag":4507,"props":24765,"children":24766},{},[24767],{"type":21,"tag":4511,"props":24768,"children":24769},{},[24770,24775],{"type":21,"tag":4515,"props":24771,"children":24772},{},[24773],{"type":26,"value":24774},"プラットフォーム",{"type":21,"tag":4515,"props":24776,"children":24777},{},[24778],{"type":26,"value":24779},"PQC の状況",{"type":21,"tag":4529,"props":24781,"children":24782},{},[24783,24796,24809,24822,24835,24848,24861,24880,24893],{"type":21,"tag":4511,"props":24784,"children":24785},{},[24786,24791],{"type":21,"tag":4536,"props":24787,"children":24788},{},[24789],{"type":26,"value":24790},"OpenSSL 3.x + oqs-provider",{"type":21,"tag":4536,"props":24792,"children":24793},{},[24794],{"type":26,"value":24795},"プラグイン経由で ML-KEM、ML-DSA、FALCON",{"type":21,"tag":4511,"props":24797,"children":24798},{},[24799,24804],{"type":21,"tag":4536,"props":24800,"children":24801},{},[24802],{"type":26,"value":24803},"AWS KMS",{"type":21,"tag":4536,"props":24805,"children":24806},{},[24807],{"type":26,"value":24808},"ML-KEM ハイブリッド TLS がプレビュー中",{"type":21,"tag":4511,"props":24810,"children":24811},{},[24812,24817],{"type":21,"tag":4536,"props":24813,"children":24814},{},[24815],{"type":26,"value":24816},"Cloudflare",{"type":21,"tag":4536,"props":24818,"children":24819},{},[24820],{"type":26,"value":24821},"X25519Kyber768 TLS (2023 年から稼働中)",{"type":21,"tag":4511,"props":24823,"children":24824},{},[24825,24830],{"type":21,"tag":4536,"props":24826,"children":24827},{},[24828],{"type":26,"value":24829},"Chrome 124+",{"type":21,"tag":4536,"props":24831,"children":24832},{},[24833],{"type":26,"value":24834},"X25519Kyber768 TLS",{"type":21,"tag":4511,"props":24836,"children":24837},{},[24838,24843],{"type":21,"tag":4536,"props":24839,"children":24840},{},[24841],{"type":26,"value":24842},"Java (BouncyCastle)",{"type":21,"tag":4536,"props":24844,"children":24845},{},[24846],{"type":26,"value":24847},"CRYSTALS-Kyber、Dilithium",{"type":21,"tag":4511,"props":24849,"children":24850},{},[24851,24856],{"type":21,"tag":4536,"props":24852,"children":24853},{},[24854],{"type":26,"value":24855},"Python (pqcrypto)",{"type":21,"tag":4536,"props":24857,"children":24858},{},[24859],{"type":26,"value":24860},"NIST の全セット",{"type":21,"tag":4511,"props":24862,"children":24863},{},[24864,24869],{"type":21,"tag":4536,"props":24865,"children":24866},{},[24867],{"type":26,"value":24868},"Go",{"type":21,"tag":4536,"props":24870,"children":24871},{},[24872,24878],{"type":21,"tag":79,"props":24873,"children":24875},{"className":24874},[],[24876],{"type":26,"value":24877},"golang.org\u002Fx\u002Fcrypto",{"type":26,"value":24879}," の PQC が開発中",{"type":21,"tag":4511,"props":24881,"children":24882},{},[24883,24888],{"type":21,"tag":4536,"props":24884,"children":24885},{},[24886],{"type":26,"value":24887},"Node.js",{"type":21,"tag":4536,"props":24889,"children":24890},{},[24891],{"type":26,"value":24892},"ネイティブモジュール経由 (OpenSSL 3.x バインディング)",{"type":21,"tag":4511,"props":24894,"children":24895},{},[24896,24901],{"type":21,"tag":4536,"props":24897,"children":24898},{},[24899],{"type":26,"value":24900},"Rust (pqcrypto crate)",{"type":21,"tag":4536,"props":24902,"children":24903},{},[24904],{"type":26,"value":24860},{"type":21,"tag":59,"props":24906,"children":24908},{"id":24907},"_1-ページ要約",[24909],{"type":26,"value":24910},"1 ページ要約",{"type":21,"tag":312,"props":24912,"children":24914},{"code":24913},"What breaks:    RSA, ECDSA, ECDH, classical DH\nWhat's safe:    AES-256, SHA-256, SHA-3, bcrypt, Argon2\nNew standards:  ML-KEM (Kyber) for key exchange\n                ML-DSA (Dilithium) for signatures\n                SLH-DSA (SPHINCS+) for conservative signing\n\nDo today:       Upgrade AES-128 → AES-256\n                Enable TLS 1.3 (forward secrecy)\n                Audit crypto inventory\n\nDo soon:        Hybrid TLS key exchange (X25519Kyber768)\n                ML-DSA for new signature systems\n\nDo in 1-5 yrs:  Full PKI migration to PQC\n                Firmware updates for IoT devices\n",[24915],{"type":21,"tag":79,"props":24916,"children":24917},{"__ignoreMap":7},[24918],{"type":26,"value":24913},{"type":21,"tag":22,"props":24920,"children":24921},{},[24922],{"type":26,"value":24923},"暗号に対する量子の脅威は現実のものであり、そのタイムラインは不確実です。だからこそ、量子コンピュータが能力を備える前の今、行動を起こすことが正しい判断なのです。",{"type":21,"tag":22,"props":24925,"children":24926},{},[24927,24932,24933,24938,24939],{"type":21,"tag":71,"props":24928,"children":24929},{},[24930],{"type":26,"value":24931},"さらに読む:",{"type":26,"value":1901},{"type":21,"tag":29,"props":24934,"children":24935},{"href":10066},[24936],{"type":26,"value":24937},"量子コンピューティング用語集 — Shor のアルゴリズム",{"type":26,"value":7549},{"type":21,"tag":29,"props":24940,"children":24941},{"href":11943},[24942],{"type":26,"value":24943},"量子コンピュータが今日できること",{"type":21,"tag":703,"props":24945,"children":24946},{},[24947],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":24949},[24950,24951,24952,24957,24962,24963],{"id":23942,"depth":335,"text":23945},{"id":24035,"depth":335,"text":24038},{"id":24142,"depth":335,"text":24145,"children":24953},[24954,24955,24956],{"id":24153,"depth":344,"text":24156},{"id":24363,"depth":344,"text":24366},{"id":24531,"depth":344,"text":24534},{"id":24588,"depth":335,"text":24591,"children":24958},[24959,24960,24961],{"id":24594,"depth":344,"text":24597},{"id":24653,"depth":344,"text":24656},{"id":24722,"depth":344,"text":24725},{"id":24758,"depth":335,"text":24758},{"id":24907,"depth":335,"text":24910},"content:ja:blog:post-quantum-cryptography-guide.md","ja\u002Fblog\u002Fpost-quantum-cryptography-guide.md","ja\u002Fblog\u002Fpost-quantum-cryptography-guide",{"_path":24968,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":24969,"description":24970,"date":24971,"author":11,"tags":24972,"readingTime":496,"body":24973,"_type":715,"_id":27511,"_source":717,"_file":27512,"_stem":27513,"_extension":720},"\u002Fja\u002Fblog\u002Freduce-shot-count","変分量子アルゴリズムでショット数を80%削減する方法","VQEおよびQAOAで必要な回路実行回数を削減するための実践的テクニック — ショット節約型オプティマイザ、勾配推定の工夫、測定のグループ化。","2026-03-01",[13,883,14,9326],{"type":18,"children":24974,"toc":27500},[24975,24980,24985,24990,24995,25116,25128,25134,25146,25446,25456,25462,25474,25797,25809,25815,25835,26028,26053,26059,26071,26220,26230,26236,26241,26574,26579,26585,26590,26763,26769,27331,27337,27466,27471,27496],{"type":21,"tag":22,"props":24976,"children":24977},{},[24978],{"type":26,"value":24979},"量子回路を実機で実行するたびに、QPU時間と費用がかかります。VQEやQAOAのような変分アルゴリズムは、収束させるために数十万回もの回路実行を必要とすることがあります。クラウドQPUでは1タスクあたり$0.075〜$0.90かかるため、これはすぐに膨れ上がります。",{"type":21,"tag":22,"props":24981,"children":24982},{},[24983],{"type":26,"value":24984},"朗報は、ほとんどのデフォルト実装が極めてショット非効率であることです。適切なテクニックを使えば、結果の品質を犠牲にすることなくショット要求量を50〜90%削減できます。",{"type":21,"tag":59,"props":24986,"children":24988},{"id":24987},"デフォルトではなぜショット数がこれほど多いのか",[24989],{"type":26,"value":24987},{"type":21,"tag":22,"props":24991,"children":24992},{},[24993],{"type":26,"value":24994},"VQEはパウリ演算子の期待値を計算します。ハミルトニアンはパウリ項の和に分解され、各項ごとに個別の回路実行が必要になります。H₂（4量子ビット）のような分子では、パウリ項は約15個です。より大きな分子ではその数は爆発的に増えます。",{"type":21,"tag":4503,"props":24996,"children":24997},{},[24998,25023],{"type":21,"tag":4507,"props":24999,"children":25000},{},[25001],{"type":21,"tag":4511,"props":25002,"children":25003},{},[25004,25009,25013,25018],{"type":21,"tag":4515,"props":25005,"children":25006},{},[25007],{"type":26,"value":25008},"分子",{"type":21,"tag":4515,"props":25010,"children":25011},{},[25012],{"type":26,"value":3567},{"type":21,"tag":4515,"props":25014,"children":25015},{},[25016],{"type":26,"value":25017},"パウリ項数",{"type":21,"tag":4515,"props":25019,"children":25020},{},[25021],{"type":26,"value":25022},"素朴なショット数\u002F反復",{"type":21,"tag":4529,"props":25024,"children":25025},{},[25026,25048,25071,25094],{"type":21,"tag":4511,"props":25027,"children":25028},{},[25029,25034,25038,25043],{"type":21,"tag":4536,"props":25030,"children":25031},{},[25032],{"type":26,"value":25033},"H₂",{"type":21,"tag":4536,"props":25035,"children":25036},{},[25037],{"type":26,"value":5969},{"type":21,"tag":4536,"props":25039,"children":25040},{},[25041],{"type":26,"value":25042},"15",{"type":21,"tag":4536,"props":25044,"children":25045},{},[25046],{"type":26,"value":25047},"15,000",{"type":21,"tag":4511,"props":25049,"children":25050},{},[25051,25056,25061,25066],{"type":21,"tag":4536,"props":25052,"children":25053},{},[25054],{"type":26,"value":25055},"LiH",{"type":21,"tag":4536,"props":25057,"children":25058},{},[25059],{"type":26,"value":25060},"12",{"type":21,"tag":4536,"props":25062,"children":25063},{},[25064],{"type":26,"value":25065},"631",{"type":21,"tag":4536,"props":25067,"children":25068},{},[25069],{"type":26,"value":25070},"631,000",{"type":21,"tag":4511,"props":25072,"children":25073},{},[25074,25079,25084,25089],{"type":21,"tag":4536,"props":25075,"children":25076},{},[25077],{"type":26,"value":25078},"BeH₂",{"type":21,"tag":4536,"props":25080,"children":25081},{},[25082],{"type":26,"value":25083},"14",{"type":21,"tag":4536,"props":25085,"children":25086},{},[25087],{"type":26,"value":25088},"666",{"type":21,"tag":4536,"props":25090,"children":25091},{},[25092],{"type":26,"value":25093},"666,000",{"type":21,"tag":4511,"props":25095,"children":25096},{},[25097,25102,25106,25111],{"type":21,"tag":4536,"props":25098,"children":25099},{},[25100],{"type":26,"value":25101},"H₂O",{"type":21,"tag":4536,"props":25103,"children":25104},{},[25105],{"type":26,"value":25083},{"type":21,"tag":4536,"props":25107,"children":25108},{},[25109],{"type":26,"value":25110},"1,086",{"type":21,"tag":4536,"props":25112,"children":25113},{},[25114],{"type":26,"value":25115},"1,086,000",{"type":21,"tag":22,"props":25117,"children":25118},{},[25119,25121,25126],{"type":26,"value":25120},"オプティマイザの反復を200回とすると、H₂Oは素朴には",{"type":21,"tag":71,"props":25122,"children":25123},{},[25124],{"type":26,"value":25125},"2億1,700万ショット",{"type":26,"value":25127},"を必要とします。以下のテクニックでこれを80〜95%削減できます。",{"type":21,"tag":59,"props":25129,"children":25131},{"id":25130},"テクニック1-測定のグループ化最大の効果",[25132],{"type":26,"value":25133},"テクニック1: 測定のグループ化（最大の効果）",{"type":21,"tag":22,"props":25135,"children":25136},{},[25137,25139,25144],{"type":26,"value":25138},"多くのパウリ項は",{"type":21,"tag":71,"props":25140,"children":25141},{},[25142],{"type":26,"value":25143},"可換",{"type":26,"value":25145},"であり、個別にではなく単一の回路実行で同時に測定できます。可換なオブザーバブルをグループ化することは、利用できる最大の削減手段です。",{"type":21,"tag":312,"props":25147,"children":25149},{"className":3626,"code":25148,"language":3628,"meta":7,"style":7},"from qiskit.primitives import StatevectorEstimator\nfrom qiskit_nature.second_q.mappers import JordanWignerMapper\nfrom qiskit_algorithms import VQE\nfrom qiskit_algorithms.optimizers import COBYLA\n\n# Qiskit automatically groups commuting Paulis in the Estimator primitive\n# This reduces shots from O(n_terms) to O(n_groups) — often 5-10x reduction\nestimator = StatevectorEstimator()\n\n# With PennyLane, use grouping explicitly:\nimport pennylane as qml\n\nH = qml.Hamiltonian(coeffs, observables)\n\n# Group commuting terms — usually reduces term count by 5-10x\ngroups = qml.grouping.group_observables(observables, grouping_type='qwc')\nprint(f\"Original terms: {len(observables)}, Groups: {len(groups)}\")\n# Original terms: 631, Groups: 68  (for LiH)\n",[25150],{"type":21,"tag":79,"props":25151,"children":25152},{"__ignoreMap":7},[25153,25173,25192,25211,25230,25237,25245,25253,25270,25277,25285,25304,25311,25328,25335,25343,25378,25438],{"type":21,"tag":322,"props":25154,"children":25155},{"class":324,"line":325},[25156,25160,25164,25168],{"type":21,"tag":322,"props":25157,"children":25158},{"style":3638},[25159],{"type":26,"value":3641},{"type":21,"tag":322,"props":25161,"children":25162},{"style":329},[25163],{"type":26,"value":22928},{"type":21,"tag":322,"props":25165,"children":25166},{"style":3638},[25167],{"type":26,"value":3651},{"type":21,"tag":322,"props":25169,"children":25170},{"style":329},[25171],{"type":26,"value":25172}," StatevectorEstimator\n",{"type":21,"tag":322,"props":25174,"children":25175},{"class":324,"line":335},[25176,25180,25184,25188],{"type":21,"tag":322,"props":25177,"children":25178},{"style":3638},[25179],{"type":26,"value":3641},{"type":21,"tag":322,"props":25181,"children":25182},{"style":329},[25183],{"type":26,"value":22844},{"type":21,"tag":322,"props":25185,"children":25186},{"style":3638},[25187],{"type":26,"value":3651},{"type":21,"tag":322,"props":25189,"children":25190},{"style":329},[25191],{"type":26,"value":22853},{"type":21,"tag":322,"props":25193,"children":25194},{"class":324,"line":344},[25195,25199,25203,25207],{"type":21,"tag":322,"props":25196,"children":25197},{"style":3638},[25198],{"type":26,"value":3641},{"type":21,"tag":322,"props":25200,"children":25201},{"style":329},[25202],{"type":26,"value":22886},{"type":21,"tag":322,"props":25204,"children":25205},{"style":3638},[25206],{"type":26,"value":3651},{"type":21,"tag":322,"props":25208,"children":25209},{"style":348},[25210],{"type":26,"value":22895},{"type":21,"tag":322,"props":25212,"children":25213},{"class":324,"line":407},[25214,25218,25222,25226],{"type":21,"tag":322,"props":25215,"children":25216},{"style":3638},[25217],{"type":26,"value":3641},{"type":21,"tag":322,"props":25219,"children":25220},{"style":329},[25221],{"type":26,"value":22907},{"type":21,"tag":322,"props":25223,"children":25224},{"style":3638},[25225],{"type":26,"value":3651},{"type":21,"tag":322,"props":25227,"children":25228},{"style":348},[25229],{"type":26,"value":22916},{"type":21,"tag":322,"props":25231,"children":25232},{"class":324,"line":431},[25233],{"type":21,"tag":322,"props":25234,"children":25235},{"emptyLinePlaceholder":3662},[25236],{"type":26,"value":3665},{"type":21,"tag":322,"props":25238,"children":25239},{"class":324,"line":453},[25240],{"type":21,"tag":322,"props":25241,"children":25242},{"style":3888},[25243],{"type":26,"value":25244},"# Qiskit automatically groups commuting Paulis in the Estimator primitive\n",{"type":21,"tag":322,"props":25246,"children":25247},{"class":324,"line":474},[25248],{"type":21,"tag":322,"props":25249,"children":25250},{"style":3888},[25251],{"type":26,"value":25252},"# This reduces shots from O(n_terms) to O(n_groups) — often 5-10x reduction\n",{"type":21,"tag":322,"props":25254,"children":25255},{"class":324,"line":496},[25256,25261,25265],{"type":21,"tag":322,"props":25257,"children":25258},{"style":329},[25259],{"type":26,"value":25260},"estimator ",{"type":21,"tag":322,"props":25262,"children":25263},{"style":3638},[25264],{"type":26,"value":3678},{"type":21,"tag":322,"props":25266,"children":25267},{"style":329},[25268],{"type":26,"value":25269}," StatevectorEstimator()\n",{"type":21,"tag":322,"props":25271,"children":25272},{"class":324,"line":16},[25273],{"type":21,"tag":322,"props":25274,"children":25275},{"emptyLinePlaceholder":3662},[25276],{"type":26,"value":3665},{"type":21,"tag":322,"props":25278,"children":25279},{"class":324,"line":3884},[25280],{"type":21,"tag":322,"props":25281,"children":25282},{"style":3888},[25283],{"type":26,"value":25284},"# With PennyLane, use grouping explicitly:\n",{"type":21,"tag":322,"props":25286,"children":25287},{"class":324,"line":3894},[25288,25292,25296,25300],{"type":21,"tag":322,"props":25289,"children":25290},{"style":3638},[25291],{"type":26,"value":3651},{"type":21,"tag":322,"props":25293,"children":25294},{"style":329},[25295],{"type":26,"value":14530},{"type":21,"tag":322,"props":25297,"children":25298},{"style":3638},[25299],{"type":26,"value":4715},{"type":21,"tag":322,"props":25301,"children":25302},{"style":329},[25303],{"type":26,"value":14539},{"type":21,"tag":322,"props":25305,"children":25306},{"class":324,"line":1034},[25307],{"type":21,"tag":322,"props":25308,"children":25309},{"emptyLinePlaceholder":3662},[25310],{"type":26,"value":3665},{"type":21,"tag":322,"props":25312,"children":25313},{"class":324,"line":3954},[25314,25319,25323],{"type":21,"tag":322,"props":25315,"children":25316},{"style":329},[25317],{"type":26,"value":25318},"H ",{"type":21,"tag":322,"props":25320,"children":25321},{"style":3638},[25322],{"type":26,"value":3678},{"type":21,"tag":322,"props":25324,"children":25325},{"style":329},[25326],{"type":26,"value":25327}," qml.Hamiltonian(coeffs, observables)\n",{"type":21,"tag":322,"props":25329,"children":25330},{"class":324,"line":4002},[25331],{"type":21,"tag":322,"props":25332,"children":25333},{"emptyLinePlaceholder":3662},[25334],{"type":26,"value":3665},{"type":21,"tag":322,"props":25336,"children":25337},{"class":324,"line":4044},[25338],{"type":21,"tag":322,"props":25339,"children":25340},{"style":3888},[25341],{"type":26,"value":25342},"# Group commuting terms — usually reduces term count by 5-10x\n",{"type":21,"tag":322,"props":25344,"children":25345},{"class":324,"line":5068},[25346,25351,25355,25360,25365,25369,25374],{"type":21,"tag":322,"props":25347,"children":25348},{"style":329},[25349],{"type":26,"value":25350},"groups ",{"type":21,"tag":322,"props":25352,"children":25353},{"style":3638},[25354],{"type":26,"value":3678},{"type":21,"tag":322,"props":25356,"children":25357},{"style":329},[25358],{"type":26,"value":25359}," qml.grouping.group_observables(observables, ",{"type":21,"tag":322,"props":25361,"children":25362},{"style":3703},[25363],{"type":26,"value":25364},"grouping_type",{"type":21,"tag":322,"props":25366,"children":25367},{"style":3638},[25368],{"type":26,"value":3678},{"type":21,"tag":322,"props":25370,"children":25371},{"style":359},[25372],{"type":26,"value":25373},"'qwc'",{"type":21,"tag":322,"props":25375,"children":25376},{"style":329},[25377],{"type":26,"value":3738},{"type":21,"tag":322,"props":25379,"children":25380},{"class":324,"line":5087},[25381,25385,25389,25393,25398,25403,25408,25412,25417,25421,25426,25430,25434],{"type":21,"tag":322,"props":25382,"children":25383},{"style":348},[25384],{"type":26,"value":3753},{"type":21,"tag":322,"props":25386,"children":25387},{"style":329},[25388],{"type":26,"value":3758},{"type":21,"tag":322,"props":25390,"children":25391},{"style":3638},[25392],{"type":26,"value":3763},{"type":21,"tag":322,"props":25394,"children":25395},{"style":359},[25396],{"type":26,"value":25397},"\"Original terms: ",{"type":21,"tag":322,"props":25399,"children":25400},{"style":348},[25401],{"type":26,"value":25402},"{len",{"type":21,"tag":322,"props":25404,"children":25405},{"style":329},[25406],{"type":26,"value":25407},"(observables)",{"type":21,"tag":322,"props":25409,"children":25410},{"style":348},[25411],{"type":26,"value":3783},{"type":21,"tag":322,"props":25413,"children":25414},{"style":359},[25415],{"type":26,"value":25416},", Groups: ",{"type":21,"tag":322,"props":25418,"children":25419},{"style":348},[25420],{"type":26,"value":25402},{"type":21,"tag":322,"props":25422,"children":25423},{"style":329},[25424],{"type":26,"value":25425},"(groups)",{"type":21,"tag":322,"props":25427,"children":25428},{"style":348},[25429],{"type":26,"value":3783},{"type":21,"tag":322,"props":25431,"children":25432},{"style":359},[25433],{"type":26,"value":3788},{"type":21,"tag":322,"props":25435,"children":25436},{"style":329},[25437],{"type":26,"value":3738},{"type":21,"tag":322,"props":25439,"children":25440},{"class":324,"line":5095},[25441],{"type":21,"tag":322,"props":25442,"children":25443},{"style":3888},[25444],{"type":26,"value":25445},"# Original terms: 631, Groups: 68  (for LiH)\n",{"type":21,"tag":22,"props":25447,"children":25448},{},[25449,25454],{"type":21,"tag":71,"props":25450,"children":25451},{},[25452],{"type":26,"value":25453},"期待される削減効果: 5〜15×",{"type":26,"value":25455}," — 典型的な化学ハミルトニアンにおいて。",{"type":21,"tag":59,"props":25457,"children":25459},{"id":25458},"テクニック2-ショット節約型オプティマイザ",[25460],{"type":26,"value":25461},"テクニック2: ショット節約型オプティマイザ",{"type":21,"tag":22,"props":25463,"children":25464},{},[25465,25467,25472],{"type":26,"value":25466},"L-BFGS-BやADAMのような古典的オプティマイザは、ノイズのない関数評価を前提としており、結果にノイズがある場合に必要以上の勾配評価を要求します。",{"type":21,"tag":71,"props":25468,"children":25469},{},[25470],{"type":26,"value":25471},"ショット節約型オプティマイザ",{"type":26,"value":25473},"は、測定の分散に基づいてショットを適応的に割り当てます。",{"type":21,"tag":312,"props":25475,"children":25477},{"className":3626,"code":25476,"language":3628,"meta":7,"style":7},"from pennylane.optimize import AdaptiveOptimizer, ShotAdaptiveOptimizer\n\ndev = qml.device(\"default.qubit\", wires=4)\n\n@qml.qnode(dev)\ndef circuit(params):\n    # ansatz\n    ...\n    return qml.expval(H)\n\n# ShotAdaptiveOptimizer: allocates more shots to high-variance directions\nopt = ShotAdaptiveOptimizer(min_shots=10)\n\nparams = np.random.uniform(-np.pi, np.pi, n_params)\nfor i in range(100):\n    params, _, shots_used = opt.step_and_cost(circuit, params)\n    print(f\"Step {i}: shots used = {shots_used}\")\n",[25478],{"type":21,"tag":79,"props":25479,"children":25480},{"__ignoreMap":7},[25481,25502,25509,25548,25555,25566,25582,25590,25598,25610,25617,25625,25658,25665,25691,25723,25740],{"type":21,"tag":322,"props":25482,"children":25483},{"class":324,"line":325},[25484,25488,25493,25497],{"type":21,"tag":322,"props":25485,"children":25486},{"style":3638},[25487],{"type":26,"value":3641},{"type":21,"tag":322,"props":25489,"children":25490},{"style":329},[25491],{"type":26,"value":25492}," pennylane.optimize ",{"type":21,"tag":322,"props":25494,"children":25495},{"style":3638},[25496],{"type":26,"value":3651},{"type":21,"tag":322,"props":25498,"children":25499},{"style":329},[25500],{"type":26,"value":25501}," AdaptiveOptimizer, ShotAdaptiveOptimizer\n",{"type":21,"tag":322,"props":25503,"children":25504},{"class":324,"line":335},[25505],{"type":21,"tag":322,"props":25506,"children":25507},{"emptyLinePlaceholder":3662},[25508],{"type":26,"value":3665},{"type":21,"tag":322,"props":25510,"children":25511},{"class":324,"line":344},[25512,25516,25520,25524,25528,25532,25536,25540,25544],{"type":21,"tag":322,"props":25513,"children":25514},{"style":329},[25515],{"type":26,"value":14663},{"type":21,"tag":322,"props":25517,"children":25518},{"style":3638},[25519],{"type":26,"value":3678},{"type":21,"tag":322,"props":25521,"children":25522},{"style":329},[25523],{"type":26,"value":14672},{"type":21,"tag":322,"props":25525,"children":25526},{"style":359},[25527],{"type":26,"value":14677},{"type":21,"tag":322,"props":25529,"children":25530},{"style":329},[25531],{"type":26,"value":367},{"type":21,"tag":322,"props":25533,"children":25534},{"style":3703},[25535],{"type":26,"value":14686},{"type":21,"tag":322,"props":25537,"children":25538},{"style":3638},[25539],{"type":26,"value":3678},{"type":21,"tag":322,"props":25541,"children":25542},{"style":348},[25543],{"type":26,"value":5969},{"type":21,"tag":322,"props":25545,"children":25546},{"style":329},[25547],{"type":26,"value":3738},{"type":21,"tag":322,"props":25549,"children":25550},{"class":324,"line":407},[25551],{"type":21,"tag":322,"props":25552,"children":25553},{"emptyLinePlaceholder":3662},[25554],{"type":26,"value":3665},{"type":21,"tag":322,"props":25556,"children":25557},{"class":324,"line":431},[25558,25562],{"type":21,"tag":322,"props":25559,"children":25560},{"style":4801},[25561],{"type":26,"value":14710},{"type":21,"tag":322,"props":25563,"children":25564},{"style":329},[25565],{"type":26,"value":14715},{"type":21,"tag":322,"props":25567,"children":25568},{"class":324,"line":453},[25569,25573,25577],{"type":21,"tag":322,"props":25570,"children":25571},{"style":3638},[25572],{"type":26,"value":4798},{"type":21,"tag":322,"props":25574,"children":25575},{"style":4801},[25576],{"type":26,"value":14727},{"type":21,"tag":322,"props":25578,"children":25579},{"style":329},[25580],{"type":26,"value":25581},"(params):\n",{"type":21,"tag":322,"props":25583,"children":25584},{"class":324,"line":474},[25585],{"type":21,"tag":322,"props":25586,"children":25587},{"style":3888},[25588],{"type":26,"value":25589},"    # ansatz\n",{"type":21,"tag":322,"props":25591,"children":25592},{"class":324,"line":496},[25593],{"type":21,"tag":322,"props":25594,"children":25595},{"style":348},[25596],{"type":26,"value":25597},"    ...\n",{"type":21,"tag":322,"props":25599,"children":25600},{"class":324,"line":16},[25601,25605],{"type":21,"tag":322,"props":25602,"children":25603},{"style":3638},[25604],{"type":26,"value":5023},{"type":21,"tag":322,"props":25606,"children":25607},{"style":329},[25608],{"type":26,"value":25609}," qml.expval(H)\n",{"type":21,"tag":322,"props":25611,"children":25612},{"class":324,"line":3884},[25613],{"type":21,"tag":322,"props":25614,"children":25615},{"emptyLinePlaceholder":3662},[25616],{"type":26,"value":3665},{"type":21,"tag":322,"props":25618,"children":25619},{"class":324,"line":3894},[25620],{"type":21,"tag":322,"props":25621,"children":25622},{"style":3888},[25623],{"type":26,"value":25624},"# ShotAdaptiveOptimizer: allocates more shots to high-variance directions\n",{"type":21,"tag":322,"props":25626,"children":25627},{"class":324,"line":1034},[25628,25632,25636,25641,25646,25650,25654],{"type":21,"tag":322,"props":25629,"children":25630},{"style":329},[25631],{"type":26,"value":15389},{"type":21,"tag":322,"props":25633,"children":25634},{"style":3638},[25635],{"type":26,"value":3678},{"type":21,"tag":322,"props":25637,"children":25638},{"style":329},[25639],{"type":26,"value":25640}," ShotAdaptiveOptimizer(",{"type":21,"tag":322,"props":25642,"children":25643},{"style":3703},[25644],{"type":26,"value":25645},"min_shots",{"type":21,"tag":322,"props":25647,"children":25648},{"style":3638},[25649],{"type":26,"value":3678},{"type":21,"tag":322,"props":25651,"children":25652},{"style":348},[25653],{"type":26,"value":20955},{"type":21,"tag":322,"props":25655,"children":25656},{"style":329},[25657],{"type":26,"value":3738},{"type":21,"tag":322,"props":25659,"children":25660},{"class":324,"line":3954},[25661],{"type":21,"tag":322,"props":25662,"children":25663},{"emptyLinePlaceholder":3662},[25664],{"type":26,"value":3665},{"type":21,"tag":322,"props":25666,"children":25667},{"class":324,"line":4002},[25668,25673,25677,25682,25686],{"type":21,"tag":322,"props":25669,"children":25670},{"style":329},[25671],{"type":26,"value":25672},"params ",{"type":21,"tag":322,"props":25674,"children":25675},{"style":3638},[25676],{"type":26,"value":3678},{"type":21,"tag":322,"props":25678,"children":25679},{"style":329},[25680],{"type":26,"value":25681}," np.random.uniform(",{"type":21,"tag":322,"props":25683,"children":25684},{"style":3638},[25685],{"type":26,"value":5285},{"type":21,"tag":322,"props":25687,"children":25688},{"style":329},[25689],{"type":26,"value":25690},"np.pi, np.pi, n_params)\n",{"type":21,"tag":322,"props":25692,"children":25693},{"class":324,"line":4044},[25694,25698,25702,25706,25710,25714,25719],{"type":21,"tag":322,"props":25695,"children":25696},{"style":3638},[25697],{"type":26,"value":3931},{"type":21,"tag":322,"props":25699,"children":25700},{"style":329},[25701],{"type":26,"value":5182},{"type":21,"tag":322,"props":25703,"children":25704},{"style":3638},[25705],{"type":26,"value":3941},{"type":21,"tag":322,"props":25707,"children":25708},{"style":348},[25709],{"type":26,"value":3946},{"type":21,"tag":322,"props":25711,"children":25712},{"style":329},[25713],{"type":26,"value":3758},{"type":21,"tag":322,"props":25715,"children":25716},{"style":348},[25717],{"type":26,"value":25718},"100",{"type":21,"tag":322,"props":25720,"children":25721},{"style":329},[25722],{"type":26,"value":4851},{"type":21,"tag":322,"props":25724,"children":25725},{"class":324,"line":5068},[25726,25731,25735],{"type":21,"tag":322,"props":25727,"children":25728},{"style":329},[25729],{"type":26,"value":25730},"    params, _, shots_used ",{"type":21,"tag":322,"props":25732,"children":25733},{"style":3638},[25734],{"type":26,"value":3678},{"type":21,"tag":322,"props":25736,"children":25737},{"style":329},[25738],{"type":26,"value":25739}," opt.step_and_cost(circuit, params)\n",{"type":21,"tag":322,"props":25741,"children":25742},{"class":324,"line":5087},[25743,25747,25751,25755,25759,25763,25767,25771,25776,25780,25785,25789,25793],{"type":21,"tag":322,"props":25744,"children":25745},{"style":348},[25746],{"type":26,"value":17663},{"type":21,"tag":322,"props":25748,"children":25749},{"style":329},[25750],{"type":26,"value":3758},{"type":21,"tag":322,"props":25752,"children":25753},{"style":3638},[25754],{"type":26,"value":3763},{"type":21,"tag":322,"props":25756,"children":25757},{"style":359},[25758],{"type":26,"value":15668},{"type":21,"tag":322,"props":25760,"children":25761},{"style":348},[25762],{"type":26,"value":3773},{"type":21,"tag":322,"props":25764,"children":25765},{"style":329},[25766],{"type":26,"value":4673},{"type":21,"tag":322,"props":25768,"children":25769},{"style":348},[25770],{"type":26,"value":3783},{"type":21,"tag":322,"props":25772,"children":25773},{"style":359},[25774],{"type":26,"value":25775},": shots used = ",{"type":21,"tag":322,"props":25777,"children":25778},{"style":348},[25779],{"type":26,"value":3773},{"type":21,"tag":322,"props":25781,"children":25782},{"style":329},[25783],{"type":26,"value":25784},"shots_used",{"type":21,"tag":322,"props":25786,"children":25787},{"style":348},[25788],{"type":26,"value":3783},{"type":21,"tag":322,"props":25790,"children":25791},{"style":359},[25792],{"type":26,"value":3788},{"type":21,"tag":322,"props":25794,"children":25795},{"style":329},[25796],{"type":26,"value":3738},{"type":21,"tag":22,"props":25798,"children":25799},{},[25800,25802,25807],{"type":26,"value":25801},"このオプティマイザは評価あたり少ないショット数から始め、勾配が不確かな場合にのみショット数を増やします。典型的なVQE実行では、固定ショットのCOBYLAと比べて",{"type":21,"tag":71,"props":25803,"children":25804},{},[25805],{"type":26,"value":25806},"5〜10×少ないショット数",{"type":26,"value":25808},"で済みます。",{"type":21,"tag":59,"props":25810,"children":25812},{"id":25811},"テクニック3-パラメータシフト勾配評価回数を減らす",[25813],{"type":26,"value":25814},"テクニック3: パラメータシフト勾配（評価回数を減らす）",{"type":21,"tag":22,"props":25816,"children":25817},{},[25818,25820,25826,25828,25833],{"type":26,"value":25819},"素朴な有限差分による勾配推定 ",{"type":21,"tag":79,"props":25821,"children":25823},{"className":25822},[],[25824],{"type":26,"value":25825},"(f(x+ε) - f(x))\u002Fε",{"type":26,"value":25827}," は、εが小さいと分散が大きく、εが大きいとバイアスが大きくなります。",{"type":21,"tag":71,"props":25829,"children":25830},{},[25831],{"type":26,"value":25832},"パラメータシフト則",{"type":26,"value":25834},"は、パラメータあたりわずか2回の回路評価で厳密な勾配を与えます。",{"type":21,"tag":312,"props":25836,"children":25838},{"className":3626,"code":25837,"language":3628,"meta":7,"style":7},"# PennyLane uses parameter-shift by default for gradients\n@qml.qnode(dev, diff_method=\"parameter-shift\")  # 2 evals per param\ndef circuit(params):\n    ...\n\n# Compare to finite-difference (requires 1 eval per param but biased):\n@qml.qnode(dev, diff_method=\"finite-diff\")  # 1 eval but approximate\n\n# For large circuits, use \"best\" — PennyLane chooses adjoint on simulator,\n# parameter-shift on hardware\n@qml.qnode(dev, diff_method=\"best\")\ndef circuit(params):\n    ...\n",[25839],{"type":21,"tag":79,"props":25840,"children":25841},{"__ignoreMap":7},[25842,25850,25885,25900,25907,25914,25922,25955,25962,25970,25978,26006,26021],{"type":21,"tag":322,"props":25843,"children":25844},{"class":324,"line":325},[25845],{"type":21,"tag":322,"props":25846,"children":25847},{"style":3888},[25848],{"type":26,"value":25849},"# PennyLane uses parameter-shift by default for gradients\n",{"type":21,"tag":322,"props":25851,"children":25852},{"class":324,"line":335},[25853,25857,25862,25867,25871,25876,25880],{"type":21,"tag":322,"props":25854,"children":25855},{"style":4801},[25856],{"type":26,"value":14710},{"type":21,"tag":322,"props":25858,"children":25859},{"style":329},[25860],{"type":26,"value":25861},"(dev, ",{"type":21,"tag":322,"props":25863,"children":25864},{"style":3703},[25865],{"type":26,"value":25866},"diff_method",{"type":21,"tag":322,"props":25868,"children":25869},{"style":3638},[25870],{"type":26,"value":3678},{"type":21,"tag":322,"props":25872,"children":25873},{"style":359},[25874],{"type":26,"value":25875},"\"parameter-shift\"",{"type":21,"tag":322,"props":25877,"children":25878},{"style":329},[25879],{"type":26,"value":19873},{"type":21,"tag":322,"props":25881,"children":25882},{"style":3888},[25883],{"type":26,"value":25884},"# 2 evals per param\n",{"type":21,"tag":322,"props":25886,"children":25887},{"class":324,"line":344},[25888,25892,25896],{"type":21,"tag":322,"props":25889,"children":25890},{"style":3638},[25891],{"type":26,"value":4798},{"type":21,"tag":322,"props":25893,"children":25894},{"style":4801},[25895],{"type":26,"value":14727},{"type":21,"tag":322,"props":25897,"children":25898},{"style":329},[25899],{"type":26,"value":25581},{"type":21,"tag":322,"props":25901,"children":25902},{"class":324,"line":407},[25903],{"type":21,"tag":322,"props":25904,"children":25905},{"style":348},[25906],{"type":26,"value":25597},{"type":21,"tag":322,"props":25908,"children":25909},{"class":324,"line":431},[25910],{"type":21,"tag":322,"props":25911,"children":25912},{"emptyLinePlaceholder":3662},[25913],{"type":26,"value":3665},{"type":21,"tag":322,"props":25915,"children":25916},{"class":324,"line":453},[25917],{"type":21,"tag":322,"props":25918,"children":25919},{"style":3888},[25920],{"type":26,"value":25921},"# Compare to finite-difference (requires 1 eval per param but biased):\n",{"type":21,"tag":322,"props":25923,"children":25924},{"class":324,"line":474},[25925,25929,25933,25937,25941,25946,25950],{"type":21,"tag":322,"props":25926,"children":25927},{"style":4801},[25928],{"type":26,"value":14710},{"type":21,"tag":322,"props":25930,"children":25931},{"style":329},[25932],{"type":26,"value":25861},{"type":21,"tag":322,"props":25934,"children":25935},{"style":3703},[25936],{"type":26,"value":25866},{"type":21,"tag":322,"props":25938,"children":25939},{"style":3638},[25940],{"type":26,"value":3678},{"type":21,"tag":322,"props":25942,"children":25943},{"style":359},[25944],{"type":26,"value":25945},"\"finite-diff\"",{"type":21,"tag":322,"props":25947,"children":25948},{"style":329},[25949],{"type":26,"value":19873},{"type":21,"tag":322,"props":25951,"children":25952},{"style":3888},[25953],{"type":26,"value":25954},"# 1 eval but approximate\n",{"type":21,"tag":322,"props":25956,"children":25957},{"class":324,"line":496},[25958],{"type":21,"tag":322,"props":25959,"children":25960},{"emptyLinePlaceholder":3662},[25961],{"type":26,"value":3665},{"type":21,"tag":322,"props":25963,"children":25964},{"class":324,"line":16},[25965],{"type":21,"tag":322,"props":25966,"children":25967},{"style":3888},[25968],{"type":26,"value":25969},"# For large circuits, use \"best\" — PennyLane chooses adjoint on simulator,\n",{"type":21,"tag":322,"props":25971,"children":25972},{"class":324,"line":3884},[25973],{"type":21,"tag":322,"props":25974,"children":25975},{"style":3888},[25976],{"type":26,"value":25977},"# parameter-shift on hardware\n",{"type":21,"tag":322,"props":25979,"children":25980},{"class":324,"line":3894},[25981,25985,25989,25993,25997,26002],{"type":21,"tag":322,"props":25982,"children":25983},{"style":4801},[25984],{"type":26,"value":14710},{"type":21,"tag":322,"props":25986,"children":25987},{"style":329},[25988],{"type":26,"value":25861},{"type":21,"tag":322,"props":25990,"children":25991},{"style":3703},[25992],{"type":26,"value":25866},{"type":21,"tag":322,"props":25994,"children":25995},{"style":3638},[25996],{"type":26,"value":3678},{"type":21,"tag":322,"props":25998,"children":25999},{"style":359},[26000],{"type":26,"value":26001},"\"best\"",{"type":21,"tag":322,"props":26003,"children":26004},{"style":329},[26005],{"type":26,"value":3738},{"type":21,"tag":322,"props":26007,"children":26008},{"class":324,"line":1034},[26009,26013,26017],{"type":21,"tag":322,"props":26010,"children":26011},{"style":3638},[26012],{"type":26,"value":4798},{"type":21,"tag":322,"props":26014,"children":26015},{"style":4801},[26016],{"type":26,"value":14727},{"type":21,"tag":322,"props":26018,"children":26019},{"style":329},[26020],{"type":26,"value":25581},{"type":21,"tag":322,"props":26022,"children":26023},{"class":324,"line":3954},[26024],{"type":21,"tag":322,"props":26025,"children":26026},{"style":348},[26027],{"type":26,"value":25597},{"type":21,"tag":22,"props":26029,"children":26030},{},[26031,26036,26038,26044,26046,26051],{"type":21,"tag":79,"props":26032,"children":26034},{"className":26033},[],[26035],{"type":26,"value":5266},{"type":26,"value":26037}," 個のパラメータに対して、パラメータシフトは勾配ステップあたり ",{"type":21,"tag":79,"props":26039,"children":26041},{"className":26040},[],[26042],{"type":26,"value":26043},"2n",{"type":26,"value":26045}," 回の評価を要します。",{"type":21,"tag":79,"props":26047,"children":26049},{"className":26048},[],[26050],{"type":26,"value":5266},{"type":26,"value":26052}," が大きい場合は、勾配を用いないオプティマイザ（COBYLA、SPSA、Nelder-Mead）を使いましょう。",{"type":21,"tag":59,"props":26054,"children":26056},{"id":26055},"テクニック4-spsa-確率的勾配推定",[26057],{"type":26,"value":26058},"テクニック4: SPSA — 確率的勾配推定",{"type":21,"tag":22,"props":26060,"children":26061},{},[26062,26064,26069],{"type":26,"value":26063},"同時摂動確率近似（SPSA）は、すべてのパラメータを同時に摂動させることで、パラメータ数に関係なくわずか",{"type":21,"tag":71,"props":26065,"children":26066},{},[26067],{"type":26,"value":26068},"2回の回路評価",{"type":26,"value":26070},"で勾配全体を推定します。",{"type":21,"tag":312,"props":26072,"children":26074},{"className":3626,"code":26073,"language":3628,"meta":7,"style":7},"from qiskit_algorithms.optimizers import SPSA\n\n# SPSA: 2 evaluations per step regardless of parameter count\n# vs parameter-shift: 2n evaluations per step\noptimizer = SPSA(maxiter=300, learning_rate=0.1, perturbation=0.05)\n\n# For 10 parameters:\n# - Parameter shift: 2×10 = 20 evals\u002Fstep × 300 steps = 6,000 total\n# - SPSA: 2 evals\u002Fstep × 300 steps = 600 total  ← 10x reduction\n",[26075],{"type":21,"tag":79,"props":26076,"children":26077},{"__ignoreMap":7},[26078,26098,26105,26113,26121,26189,26196,26204,26212],{"type":21,"tag":322,"props":26079,"children":26080},{"class":324,"line":325},[26081,26085,26089,26093],{"type":21,"tag":322,"props":26082,"children":26083},{"style":3638},[26084],{"type":26,"value":3641},{"type":21,"tag":322,"props":26086,"children":26087},{"style":329},[26088],{"type":26,"value":22907},{"type":21,"tag":322,"props":26090,"children":26091},{"style":3638},[26092],{"type":26,"value":3651},{"type":21,"tag":322,"props":26094,"children":26095},{"style":348},[26096],{"type":26,"value":26097}," SPSA\n",{"type":21,"tag":322,"props":26099,"children":26100},{"class":324,"line":335},[26101],{"type":21,"tag":322,"props":26102,"children":26103},{"emptyLinePlaceholder":3662},[26104],{"type":26,"value":3665},{"type":21,"tag":322,"props":26106,"children":26107},{"class":324,"line":344},[26108],{"type":21,"tag":322,"props":26109,"children":26110},{"style":3888},[26111],{"type":26,"value":26112},"# SPSA: 2 evaluations per step regardless of parameter count\n",{"type":21,"tag":322,"props":26114,"children":26115},{"class":324,"line":407},[26116],{"type":21,"tag":322,"props":26117,"children":26118},{"style":3888},[26119],{"type":26,"value":26120},"# vs parameter-shift: 2n evaluations per step\n",{"type":21,"tag":322,"props":26122,"children":26123},{"class":324,"line":431},[26124,26129,26133,26138,26142,26146,26151,26155,26160,26164,26168,26172,26177,26181,26185],{"type":21,"tag":322,"props":26125,"children":26126},{"style":329},[26127],{"type":26,"value":26128},"optimizer ",{"type":21,"tag":322,"props":26130,"children":26131},{"style":3638},[26132],{"type":26,"value":3678},{"type":21,"tag":322,"props":26134,"children":26135},{"style":329},[26136],{"type":26,"value":26137}," SPSA(",{"type":21,"tag":322,"props":26139,"children":26140},{"style":3703},[26141],{"type":26,"value":617},{"type":21,"tag":322,"props":26143,"children":26144},{"style":3638},[26145],{"type":26,"value":3678},{"type":21,"tag":322,"props":26147,"children":26148},{"style":348},[26149],{"type":26,"value":26150},"300",{"type":21,"tag":322,"props":26152,"children":26153},{"style":329},[26154],{"type":26,"value":367},{"type":21,"tag":322,"props":26156,"children":26157},{"style":3703},[26158],{"type":26,"value":26159},"learning_rate",{"type":21,"tag":322,"props":26161,"children":26162},{"style":3638},[26163],{"type":26,"value":3678},{"type":21,"tag":322,"props":26165,"children":26166},{"style":348},[26167],{"type":26,"value":15310},{"type":21,"tag":322,"props":26169,"children":26170},{"style":329},[26171],{"type":26,"value":367},{"type":21,"tag":322,"props":26173,"children":26174},{"style":3703},[26175],{"type":26,"value":26176},"perturbation",{"type":21,"tag":322,"props":26178,"children":26179},{"style":3638},[26180],{"type":26,"value":3678},{"type":21,"tag":322,"props":26182,"children":26183},{"style":348},[26184],{"type":26,"value":4846},{"type":21,"tag":322,"props":26186,"children":26187},{"style":329},[26188],{"type":26,"value":3738},{"type":21,"tag":322,"props":26190,"children":26191},{"class":324,"line":453},[26192],{"type":21,"tag":322,"props":26193,"children":26194},{"emptyLinePlaceholder":3662},[26195],{"type":26,"value":3665},{"type":21,"tag":322,"props":26197,"children":26198},{"class":324,"line":474},[26199],{"type":21,"tag":322,"props":26200,"children":26201},{"style":3888},[26202],{"type":26,"value":26203},"# For 10 parameters:\n",{"type":21,"tag":322,"props":26205,"children":26206},{"class":324,"line":496},[26207],{"type":21,"tag":322,"props":26208,"children":26209},{"style":3888},[26210],{"type":26,"value":26211},"# - Parameter shift: 2×10 = 20 evals\u002Fstep × 300 steps = 6,000 total\n",{"type":21,"tag":322,"props":26213,"children":26214},{"class":324,"line":16},[26215],{"type":21,"tag":322,"props":26216,"children":26217},{"style":3888},[26218],{"type":26,"value":26219},"# - SPSA: 2 evals\u002Fstep × 300 steps = 600 total  ← 10x reduction\n",{"type":21,"tag":22,"props":26221,"children":26222},{},[26223,26228],{"type":21,"tag":71,"props":26224,"children":26225},{},[26226],{"type":26,"value":26227},"最適な場面:",{"type":26,"value":26229}," 回路のパラメータ数が多い場合（> 10）。トレードオフとしてステップあたりの収束は遅くなりますが、総ショット数は少なくなります。",{"type":21,"tag":59,"props":26231,"children":26233},{"id":26232},"テクニック5-早期終了-分散しきい値",[26234],{"type":26,"value":26235},"テクニック5: 早期終了 + 分散しきい値",{"type":21,"tag":22,"props":26237,"children":26238},{},[26239],{"type":26,"value":26240},"結果がすでに十分に良い場合、回路を最後まで実行しないようにします。",{"type":21,"tag":312,"props":26242,"children":26244},{"className":3626,"code":26243,"language":3628,"meta":7,"style":7},"from pennylane.optimize import AdamOptimizer\nimport numpy as np\n\nopt = AdamOptimizer(stepsize=0.02)\nparams = init_params.copy()\nprev_energy = float('inf')\n\nfor step in range(max_steps):\n    params, energy = opt.step_and_cost(circuit, params)\n\n    # Stop when change is below shot-noise floor\n    variance = 1.0 \u002F np.sqrt(shots_per_eval)  # shot noise floor\n    if abs(energy - prev_energy) \u003C variance:\n        print(f\"Converged at step {step}\")\n        break\n\n    prev_energy = energy\n",[26245],{"type":21,"tag":79,"props":26246,"children":26247},{"__ignoreMap":7},[26248,26268,26287,26294,26326,26342,26372,26379,26403,26419,26426,26434,26466,26502,26542,26550,26557],{"type":21,"tag":322,"props":26249,"children":26250},{"class":324,"line":325},[26251,26255,26259,26263],{"type":21,"tag":322,"props":26252,"children":26253},{"style":3638},[26254],{"type":26,"value":3641},{"type":21,"tag":322,"props":26256,"children":26257},{"style":329},[26258],{"type":26,"value":25492},{"type":21,"tag":322,"props":26260,"children":26261},{"style":3638},[26262],{"type":26,"value":3651},{"type":21,"tag":322,"props":26264,"children":26265},{"style":329},[26266],{"type":26,"value":26267}," AdamOptimizer\n",{"type":21,"tag":322,"props":26269,"children":26270},{"class":324,"line":335},[26271,26275,26279,26283],{"type":21,"tag":322,"props":26272,"children":26273},{"style":3638},[26274],{"type":26,"value":3651},{"type":21,"tag":322,"props":26276,"children":26277},{"style":329},[26278],{"type":26,"value":4710},{"type":21,"tag":322,"props":26280,"children":26281},{"style":3638},[26282],{"type":26,"value":4715},{"type":21,"tag":322,"props":26284,"children":26285},{"style":329},[26286],{"type":26,"value":4720},{"type":21,"tag":322,"props":26288,"children":26289},{"class":324,"line":344},[26290],{"type":21,"tag":322,"props":26291,"children":26292},{"emptyLinePlaceholder":3662},[26293],{"type":26,"value":3665},{"type":21,"tag":322,"props":26295,"children":26296},{"class":324,"line":407},[26297,26301,26305,26310,26314,26318,26322],{"type":21,"tag":322,"props":26298,"children":26299},{"style":329},[26300],{"type":26,"value":15389},{"type":21,"tag":322,"props":26302,"children":26303},{"style":3638},[26304],{"type":26,"value":3678},{"type":21,"tag":322,"props":26306,"children":26307},{"style":329},[26308],{"type":26,"value":26309}," AdamOptimizer(",{"type":21,"tag":322,"props":26311,"children":26312},{"style":3703},[26313],{"type":26,"value":15403},{"type":21,"tag":322,"props":26315,"children":26316},{"style":3638},[26317],{"type":26,"value":3678},{"type":21,"tag":322,"props":26319,"children":26320},{"style":348},[26321],{"type":26,"value":4832},{"type":21,"tag":322,"props":26323,"children":26324},{"style":329},[26325],{"type":26,"value":3738},{"type":21,"tag":322,"props":26327,"children":26328},{"class":324,"line":431},[26329,26333,26337],{"type":21,"tag":322,"props":26330,"children":26331},{"style":329},[26332],{"type":26,"value":25672},{"type":21,"tag":322,"props":26334,"children":26335},{"style":3638},[26336],{"type":26,"value":3678},{"type":21,"tag":322,"props":26338,"children":26339},{"style":329},[26340],{"type":26,"value":26341}," init_params.copy()\n",{"type":21,"tag":322,"props":26343,"children":26344},{"class":324,"line":453},[26345,26350,26354,26359,26363,26368],{"type":21,"tag":322,"props":26346,"children":26347},{"style":329},[26348],{"type":26,"value":26349},"prev_energy ",{"type":21,"tag":322,"props":26351,"children":26352},{"style":3638},[26353],{"type":26,"value":3678},{"type":21,"tag":322,"props":26355,"children":26356},{"style":348},[26357],{"type":26,"value":26358}," float",{"type":21,"tag":322,"props":26360,"children":26361},{"style":329},[26362],{"type":26,"value":3758},{"type":21,"tag":322,"props":26364,"children":26365},{"style":359},[26366],{"type":26,"value":26367},"'inf'",{"type":21,"tag":322,"props":26369,"children":26370},{"style":329},[26371],{"type":26,"value":3738},{"type":21,"tag":322,"props":26373,"children":26374},{"class":324,"line":474},[26375],{"type":21,"tag":322,"props":26376,"children":26377},{"emptyLinePlaceholder":3662},[26378],{"type":26,"value":3665},{"type":21,"tag":322,"props":26380,"children":26381},{"class":324,"line":496},[26382,26386,26390,26394,26398],{"type":21,"tag":322,"props":26383,"children":26384},{"style":3638},[26385],{"type":26,"value":3931},{"type":21,"tag":322,"props":26387,"children":26388},{"style":329},[26389],{"type":26,"value":15451},{"type":21,"tag":322,"props":26391,"children":26392},{"style":3638},[26393],{"type":26,"value":3941},{"type":21,"tag":322,"props":26395,"children":26396},{"style":348},[26397],{"type":26,"value":3946},{"type":21,"tag":322,"props":26399,"children":26400},{"style":329},[26401],{"type":26,"value":26402},"(max_steps):\n",{"type":21,"tag":322,"props":26404,"children":26405},{"class":324,"line":16},[26406,26411,26415],{"type":21,"tag":322,"props":26407,"children":26408},{"style":329},[26409],{"type":26,"value":26410},"    params, energy ",{"type":21,"tag":322,"props":26412,"children":26413},{"style":3638},[26414],{"type":26,"value":3678},{"type":21,"tag":322,"props":26416,"children":26417},{"style":329},[26418],{"type":26,"value":25739},{"type":21,"tag":322,"props":26420,"children":26421},{"class":324,"line":3884},[26422],{"type":21,"tag":322,"props":26423,"children":26424},{"emptyLinePlaceholder":3662},[26425],{"type":26,"value":3665},{"type":21,"tag":322,"props":26427,"children":26428},{"class":324,"line":3894},[26429],{"type":21,"tag":322,"props":26430,"children":26431},{"style":3888},[26432],{"type":26,"value":26433},"    # Stop when change is below shot-noise floor\n",{"type":21,"tag":322,"props":26435,"children":26436},{"class":324,"line":1034},[26437,26442,26446,26451,26456,26461],{"type":21,"tag":322,"props":26438,"children":26439},{"style":329},[26440],{"type":26,"value":26441},"    variance ",{"type":21,"tag":322,"props":26443,"children":26444},{"style":3638},[26445],{"type":26,"value":3678},{"type":21,"tag":322,"props":26447,"children":26448},{"style":348},[26449],{"type":26,"value":26450}," 1.0",{"type":21,"tag":322,"props":26452,"children":26453},{"style":3638},[26454],{"type":26,"value":26455}," \u002F",{"type":21,"tag":322,"props":26457,"children":26458},{"style":329},[26459],{"type":26,"value":26460}," np.sqrt(shots_per_eval)  ",{"type":21,"tag":322,"props":26462,"children":26463},{"style":3888},[26464],{"type":26,"value":26465},"# shot noise floor\n",{"type":21,"tag":322,"props":26467,"children":26468},{"class":324,"line":3954},[26469,26473,26478,26483,26487,26492,26497],{"type":21,"tag":322,"props":26470,"children":26471},{"style":3638},[26472],{"type":26,"value":12384},{"type":21,"tag":322,"props":26474,"children":26475},{"style":348},[26476],{"type":26,"value":26477}," abs",{"type":21,"tag":322,"props":26479,"children":26480},{"style":329},[26481],{"type":26,"value":26482},"(energy ",{"type":21,"tag":322,"props":26484,"children":26485},{"style":3638},[26486],{"type":26,"value":5285},{"type":21,"tag":322,"props":26488,"children":26489},{"style":329},[26490],{"type":26,"value":26491}," prev_energy) ",{"type":21,"tag":322,"props":26493,"children":26494},{"style":3638},[26495],{"type":26,"value":26496},"\u003C",{"type":21,"tag":322,"props":26498,"children":26499},{"style":329},[26500],{"type":26,"value":26501}," variance:\n",{"type":21,"tag":322,"props":26503,"children":26504},{"class":324,"line":4002},[26505,26509,26513,26517,26522,26526,26530,26534,26538],{"type":21,"tag":322,"props":26506,"children":26507},{"style":348},[26508],{"type":26,"value":13330},{"type":21,"tag":322,"props":26510,"children":26511},{"style":329},[26512],{"type":26,"value":3758},{"type":21,"tag":322,"props":26514,"children":26515},{"style":3638},[26516],{"type":26,"value":3763},{"type":21,"tag":322,"props":26518,"children":26519},{"style":359},[26520],{"type":26,"value":26521},"\"Converged at step ",{"type":21,"tag":322,"props":26523,"children":26524},{"style":348},[26525],{"type":26,"value":3773},{"type":21,"tag":322,"props":26527,"children":26528},{"style":329},[26529],{"type":26,"value":15677},{"type":21,"tag":322,"props":26531,"children":26532},{"style":348},[26533],{"type":26,"value":3783},{"type":21,"tag":322,"props":26535,"children":26536},{"style":359},[26537],{"type":26,"value":3788},{"type":21,"tag":322,"props":26539,"children":26540},{"style":329},[26541],{"type":26,"value":3738},{"type":21,"tag":322,"props":26543,"children":26544},{"class":324,"line":4044},[26545],{"type":21,"tag":322,"props":26546,"children":26547},{"style":3638},[26548],{"type":26,"value":26549},"        break\n",{"type":21,"tag":322,"props":26551,"children":26552},{"class":324,"line":5068},[26553],{"type":21,"tag":322,"props":26554,"children":26555},{"emptyLinePlaceholder":3662},[26556],{"type":26,"value":3665},{"type":21,"tag":322,"props":26558,"children":26559},{"class":324,"line":5087},[26560,26565,26569],{"type":21,"tag":322,"props":26561,"children":26562},{"style":329},[26563],{"type":26,"value":26564},"    prev_energy ",{"type":21,"tag":322,"props":26566,"children":26567},{"style":3638},[26568],{"type":26,"value":3678},{"type":21,"tag":322,"props":26570,"children":26571},{"style":329},[26572],{"type":26,"value":26573}," energy\n",{"type":21,"tag":22,"props":26575,"children":26576},{},[26577],{"type":26,"value":26578},"多くのVQE実行は早期に頭打ちになります。それ以上続けても、ノイズの変動に対してショットを浪費するだけです。",{"type":21,"tag":59,"props":26580,"children":26582},{"id":26581},"テクニック6-ウォームスタート",[26583],{"type":26,"value":26584},"テクニック6: ウォームスタート",{"type":21,"tag":22,"props":26586,"children":26587},{},[26588],{"type":26,"value":26589},"QAOAやVQEのパラメータをランダムではなく、関連する古典解から初期化します。",{"type":21,"tag":312,"props":26591,"children":26593},{"className":3626,"code":26592,"language":3628,"meta":7,"style":7},"# For QAOA on Max-Cut: warm start from a greedy classical solution\nimport networkx as nx\n\nG = nx.from_edgelist(edges)\nclassical_cut = nx.algorithms.approximation.one_exchange(G)\n\n# Map classical solution to initial QAOA angles\n# γ₀ ≈ π\u002F4 for a good cut, β₀ ≈ π\u002F8\ninit_gamma = [np.pi \u002F 4]\ninit_beta = [np.pi \u002F 8]\n\n# Warm-started QAOA typically converges in 30-50% fewer iterations\n",[26594],{"type":21,"tag":79,"props":26595,"children":26596},{"__ignoreMap":7},[26597,26605,26626,26633,26650,26667,26674,26682,26690,26719,26748,26755],{"type":21,"tag":322,"props":26598,"children":26599},{"class":324,"line":325},[26600],{"type":21,"tag":322,"props":26601,"children":26602},{"style":3888},[26603],{"type":26,"value":26604},"# For QAOA on Max-Cut: warm start from a greedy classical solution\n",{"type":21,"tag":322,"props":26606,"children":26607},{"class":324,"line":335},[26608,26612,26617,26621],{"type":21,"tag":322,"props":26609,"children":26610},{"style":3638},[26611],{"type":26,"value":3651},{"type":21,"tag":322,"props":26613,"children":26614},{"style":329},[26615],{"type":26,"value":26616}," networkx ",{"type":21,"tag":322,"props":26618,"children":26619},{"style":3638},[26620],{"type":26,"value":4715},{"type":21,"tag":322,"props":26622,"children":26623},{"style":329},[26624],{"type":26,"value":26625}," nx\n",{"type":21,"tag":322,"props":26627,"children":26628},{"class":324,"line":344},[26629],{"type":21,"tag":322,"props":26630,"children":26631},{"emptyLinePlaceholder":3662},[26632],{"type":26,"value":3665},{"type":21,"tag":322,"props":26634,"children":26635},{"class":324,"line":407},[26636,26641,26645],{"type":21,"tag":322,"props":26637,"children":26638},{"style":329},[26639],{"type":26,"value":26640},"G ",{"type":21,"tag":322,"props":26642,"children":26643},{"style":3638},[26644],{"type":26,"value":3678},{"type":21,"tag":322,"props":26646,"children":26647},{"style":329},[26648],{"type":26,"value":26649}," nx.from_edgelist(edges)\n",{"type":21,"tag":322,"props":26651,"children":26652},{"class":324,"line":431},[26653,26658,26662],{"type":21,"tag":322,"props":26654,"children":26655},{"style":329},[26656],{"type":26,"value":26657},"classical_cut ",{"type":21,"tag":322,"props":26659,"children":26660},{"style":3638},[26661],{"type":26,"value":3678},{"type":21,"tag":322,"props":26663,"children":26664},{"style":329},[26665],{"type":26,"value":26666}," nx.algorithms.approximation.one_exchange(G)\n",{"type":21,"tag":322,"props":26668,"children":26669},{"class":324,"line":453},[26670],{"type":21,"tag":322,"props":26671,"children":26672},{"emptyLinePlaceholder":3662},[26673],{"type":26,"value":3665},{"type":21,"tag":322,"props":26675,"children":26676},{"class":324,"line":474},[26677],{"type":21,"tag":322,"props":26678,"children":26679},{"style":3888},[26680],{"type":26,"value":26681},"# Map classical solution to initial QAOA angles\n",{"type":21,"tag":322,"props":26683,"children":26684},{"class":324,"line":496},[26685],{"type":21,"tag":322,"props":26686,"children":26687},{"style":3888},[26688],{"type":26,"value":26689},"# γ₀ ≈ π\u002F4 for a good cut, β₀ ≈ π\u002F8\n",{"type":21,"tag":322,"props":26691,"children":26692},{"class":324,"line":16},[26693,26698,26702,26707,26711,26715],{"type":21,"tag":322,"props":26694,"children":26695},{"style":329},[26696],{"type":26,"value":26697},"init_gamma ",{"type":21,"tag":322,"props":26699,"children":26700},{"style":3638},[26701],{"type":26,"value":3678},{"type":21,"tag":322,"props":26703,"children":26704},{"style":329},[26705],{"type":26,"value":26706}," [np.pi ",{"type":21,"tag":322,"props":26708,"children":26709},{"style":3638},[26710],{"type":26,"value":174},{"type":21,"tag":322,"props":26712,"children":26713},{"style":348},[26714],{"type":26,"value":8777},{"type":21,"tag":322,"props":26716,"children":26717},{"style":329},[26718],{"type":26,"value":7124},{"type":21,"tag":322,"props":26720,"children":26721},{"class":324,"line":3884},[26722,26727,26731,26735,26739,26744],{"type":21,"tag":322,"props":26723,"children":26724},{"style":329},[26725],{"type":26,"value":26726},"init_beta ",{"type":21,"tag":322,"props":26728,"children":26729},{"style":3638},[26730],{"type":26,"value":3678},{"type":21,"tag":322,"props":26732,"children":26733},{"style":329},[26734],{"type":26,"value":26706},{"type":21,"tag":322,"props":26736,"children":26737},{"style":3638},[26738],{"type":26,"value":174},{"type":21,"tag":322,"props":26740,"children":26741},{"style":348},[26742],{"type":26,"value":26743}," 8",{"type":21,"tag":322,"props":26745,"children":26746},{"style":329},[26747],{"type":26,"value":7124},{"type":21,"tag":322,"props":26749,"children":26750},{"class":324,"line":3894},[26751],{"type":21,"tag":322,"props":26752,"children":26753},{"emptyLinePlaceholder":3662},[26754],{"type":26,"value":3665},{"type":21,"tag":322,"props":26756,"children":26757},{"class":324,"line":1034},[26758],{"type":21,"tag":322,"props":26759,"children":26760},{"style":3888},[26761],{"type":26,"value":26762},"# Warm-started QAOA typically converges in 30-50% fewer iterations\n",{"type":21,"tag":59,"props":26764,"children":26766},{"id":26765},"すべてを組み合わせる-実践的なvqeテンプレート",[26767],{"type":26,"value":26768},"すべてを組み合わせる: 実践的なVQEテンプレート",{"type":21,"tag":312,"props":26770,"children":26772},{"className":3626,"code":26771,"language":3628,"meta":7,"style":7},"import pennylane as qml\nimport numpy as np\nfrom pennylane.optimize import ShotAdaptiveOptimizer\n\ndev = qml.device(\"default.qubit\", wires=n_qubits, shots=512)\n\n# 1. Group commuting terms (5-10x reduction in circuit count)\ngrouped_H = qml.Hamiltonian(*qml.grouping.group_observables(H))\n\n@qml.qnode(dev, diff_method=\"parameter-shift\")\ndef ansatz(params):\n    # Hardware-efficient ansatz\n    for i in range(n_qubits):\n        qml.RY(params[i], wires=i)\n    for i in range(n_qubits - 1):\n        qml.CNOT(wires=[i, i + 1])\n    return qml.expval(grouped_H)\n\n# 2. Use shot-adaptive optimizer\nopt = ShotAdaptiveOptimizer(min_shots=50)\n\n# 3. Warm start\nparams = warm_start_params(H)\n\n# 4. Run with early stopping\nfor step in range(300):\n    params, energy, shots = opt.step_and_cost(ansatz, params)\n    if check_convergence(energy, shots):\n        break\n\nprint(f\"Ground state energy: {energy:.4f} Ha\")\n",[26773],{"type":21,"tag":79,"props":26774,"children":26775},{"__ignoreMap":7},[26776,26795,26814,26834,26841,26894,26901,26909,26935,26942,26969,26985,26993,27017,27038,27074,27107,27119,27126,27134,27166,27173,27181,27197,27204,27212,27243,27260,27272,27279,27286],{"type":21,"tag":322,"props":26777,"children":26778},{"class":324,"line":325},[26779,26783,26787,26791],{"type":21,"tag":322,"props":26780,"children":26781},{"style":3638},[26782],{"type":26,"value":3651},{"type":21,"tag":322,"props":26784,"children":26785},{"style":329},[26786],{"type":26,"value":14530},{"type":21,"tag":322,"props":26788,"children":26789},{"style":3638},[26790],{"type":26,"value":4715},{"type":21,"tag":322,"props":26792,"children":26793},{"style":329},[26794],{"type":26,"value":14539},{"type":21,"tag":322,"props":26796,"children":26797},{"class":324,"line":335},[26798,26802,26806,26810],{"type":21,"tag":322,"props":26799,"children":26800},{"style":3638},[26801],{"type":26,"value":3651},{"type":21,"tag":322,"props":26803,"children":26804},{"style":329},[26805],{"type":26,"value":4710},{"type":21,"tag":322,"props":26807,"children":26808},{"style":3638},[26809],{"type":26,"value":4715},{"type":21,"tag":322,"props":26811,"children":26812},{"style":329},[26813],{"type":26,"value":4720},{"type":21,"tag":322,"props":26815,"children":26816},{"class":324,"line":344},[26817,26821,26825,26829],{"type":21,"tag":322,"props":26818,"children":26819},{"style":3638},[26820],{"type":26,"value":3641},{"type":21,"tag":322,"props":26822,"children":26823},{"style":329},[26824],{"type":26,"value":25492},{"type":21,"tag":322,"props":26826,"children":26827},{"style":3638},[26828],{"type":26,"value":3651},{"type":21,"tag":322,"props":26830,"children":26831},{"style":329},[26832],{"type":26,"value":26833}," ShotAdaptiveOptimizer\n",{"type":21,"tag":322,"props":26835,"children":26836},{"class":324,"line":407},[26837],{"type":21,"tag":322,"props":26838,"children":26839},{"emptyLinePlaceholder":3662},[26840],{"type":26,"value":3665},{"type":21,"tag":322,"props":26842,"children":26843},{"class":324,"line":431},[26844,26848,26852,26856,26860,26864,26868,26872,26877,26881,26885,26890],{"type":21,"tag":322,"props":26845,"children":26846},{"style":329},[26847],{"type":26,"value":14663},{"type":21,"tag":322,"props":26849,"children":26850},{"style":3638},[26851],{"type":26,"value":3678},{"type":21,"tag":322,"props":26853,"children":26854},{"style":329},[26855],{"type":26,"value":14672},{"type":21,"tag":322,"props":26857,"children":26858},{"style":359},[26859],{"type":26,"value":14677},{"type":21,"tag":322,"props":26861,"children":26862},{"style":329},[26863],{"type":26,"value":367},{"type":21,"tag":322,"props":26865,"children":26866},{"style":3703},[26867],{"type":26,"value":14686},{"type":21,"tag":322,"props":26869,"children":26870},{"style":3638},[26871],{"type":26,"value":3678},{"type":21,"tag":322,"props":26873,"children":26874},{"style":329},[26875],{"type":26,"value":26876},"n_qubits, ",{"type":21,"tag":322,"props":26878,"children":26879},{"style":3703},[26880],{"type":26,"value":4560},{"type":21,"tag":322,"props":26882,"children":26883},{"style":3638},[26884],{"type":26,"value":3678},{"type":21,"tag":322,"props":26886,"children":26887},{"style":348},[26888],{"type":26,"value":26889},"512",{"type":21,"tag":322,"props":26891,"children":26892},{"style":329},[26893],{"type":26,"value":3738},{"type":21,"tag":322,"props":26895,"children":26896},{"class":324,"line":453},[26897],{"type":21,"tag":322,"props":26898,"children":26899},{"emptyLinePlaceholder":3662},[26900],{"type":26,"value":3665},{"type":21,"tag":322,"props":26902,"children":26903},{"class":324,"line":474},[26904],{"type":21,"tag":322,"props":26905,"children":26906},{"style":3888},[26907],{"type":26,"value":26908},"# 1. Group commuting terms (5-10x reduction in circuit count)\n",{"type":21,"tag":322,"props":26910,"children":26911},{"class":324,"line":496},[26912,26917,26921,26926,26930],{"type":21,"tag":322,"props":26913,"children":26914},{"style":329},[26915],{"type":26,"value":26916},"grouped_H ",{"type":21,"tag":322,"props":26918,"children":26919},{"style":3638},[26920],{"type":26,"value":3678},{"type":21,"tag":322,"props":26922,"children":26923},{"style":329},[26924],{"type":26,"value":26925}," qml.Hamiltonian(",{"type":21,"tag":322,"props":26927,"children":26928},{"style":3638},[26929],{"type":26,"value":6830},{"type":21,"tag":322,"props":26931,"children":26932},{"style":329},[26933],{"type":26,"value":26934},"qml.grouping.group_observables(H))\n",{"type":21,"tag":322,"props":26936,"children":26937},{"class":324,"line":16},[26938],{"type":21,"tag":322,"props":26939,"children":26940},{"emptyLinePlaceholder":3662},[26941],{"type":26,"value":3665},{"type":21,"tag":322,"props":26943,"children":26944},{"class":324,"line":3884},[26945,26949,26953,26957,26961,26965],{"type":21,"tag":322,"props":26946,"children":26947},{"style":4801},[26948],{"type":26,"value":14710},{"type":21,"tag":322,"props":26950,"children":26951},{"style":329},[26952],{"type":26,"value":25861},{"type":21,"tag":322,"props":26954,"children":26955},{"style":3703},[26956],{"type":26,"value":25866},{"type":21,"tag":322,"props":26958,"children":26959},{"style":3638},[26960],{"type":26,"value":3678},{"type":21,"tag":322,"props":26962,"children":26963},{"style":359},[26964],{"type":26,"value":25875},{"type":21,"tag":322,"props":26966,"children":26967},{"style":329},[26968],{"type":26,"value":3738},{"type":21,"tag":322,"props":26970,"children":26971},{"class":324,"line":3894},[26972,26976,26981],{"type":21,"tag":322,"props":26973,"children":26974},{"style":3638},[26975],{"type":26,"value":4798},{"type":21,"tag":322,"props":26977,"children":26978},{"style":4801},[26979],{"type":26,"value":26980}," ansatz",{"type":21,"tag":322,"props":26982,"children":26983},{"style":329},[26984],{"type":26,"value":25581},{"type":21,"tag":322,"props":26986,"children":26987},{"class":324,"line":1034},[26988],{"type":21,"tag":322,"props":26989,"children":26990},{"style":3888},[26991],{"type":26,"value":26992},"    # Hardware-efficient ansatz\n",{"type":21,"tag":322,"props":26994,"children":26995},{"class":324,"line":3954},[26996,27000,27004,27008,27012],{"type":21,"tag":322,"props":26997,"children":26998},{"style":3638},[26999],{"type":26,"value":5177},{"type":21,"tag":322,"props":27001,"children":27002},{"style":329},[27003],{"type":26,"value":5182},{"type":21,"tag":322,"props":27005,"children":27006},{"style":3638},[27007],{"type":26,"value":3941},{"type":21,"tag":322,"props":27009,"children":27010},{"style":348},[27011],{"type":26,"value":3946},{"type":21,"tag":322,"props":27013,"children":27014},{"style":329},[27015],{"type":26,"value":27016},"(n_qubits):\n",{"type":21,"tag":322,"props":27018,"children":27019},{"class":324,"line":4002},[27020,27025,27029,27033],{"type":21,"tag":322,"props":27021,"children":27022},{"style":329},[27023],{"type":26,"value":27024},"        qml.RY(params[i], ",{"type":21,"tag":322,"props":27026,"children":27027},{"style":3703},[27028],{"type":26,"value":14686},{"type":21,"tag":322,"props":27030,"children":27031},{"style":3638},[27032],{"type":26,"value":3678},{"type":21,"tag":322,"props":27034,"children":27035},{"style":329},[27036],{"type":26,"value":27037},"i)\n",{"type":21,"tag":322,"props":27039,"children":27040},{"class":324,"line":4044},[27041,27045,27049,27053,27057,27062,27066,27070],{"type":21,"tag":322,"props":27042,"children":27043},{"style":3638},[27044],{"type":26,"value":5177},{"type":21,"tag":322,"props":27046,"children":27047},{"style":329},[27048],{"type":26,"value":5182},{"type":21,"tag":322,"props":27050,"children":27051},{"style":3638},[27052],{"type":26,"value":3941},{"type":21,"tag":322,"props":27054,"children":27055},{"style":348},[27056],{"type":26,"value":3946},{"type":21,"tag":322,"props":27058,"children":27059},{"style":329},[27060],{"type":26,"value":27061},"(n_qubits ",{"type":21,"tag":322,"props":27063,"children":27064},{"style":3638},[27065],{"type":26,"value":5285},{"type":21,"tag":322,"props":27067,"children":27068},{"style":348},[27069],{"type":26,"value":6536},{"type":21,"tag":322,"props":27071,"children":27072},{"style":329},[27073],{"type":26,"value":4851},{"type":21,"tag":322,"props":27075,"children":27076},{"class":324,"line":5068},[27077,27082,27086,27090,27095,27099,27103],{"type":21,"tag":322,"props":27078,"children":27079},{"style":329},[27080],{"type":26,"value":27081},"        qml.CNOT(",{"type":21,"tag":322,"props":27083,"children":27084},{"style":3703},[27085],{"type":26,"value":14686},{"type":21,"tag":322,"props":27087,"children":27088},{"style":3638},[27089],{"type":26,"value":3678},{"type":21,"tag":322,"props":27091,"children":27092},{"style":329},[27093],{"type":26,"value":27094},"[i, i ",{"type":21,"tag":322,"props":27096,"children":27097},{"style":3638},[27098],{"type":26,"value":6965},{"type":21,"tag":322,"props":27100,"children":27101},{"style":348},[27102],{"type":26,"value":6536},{"type":21,"tag":322,"props":27104,"children":27105},{"style":329},[27106],{"type":26,"value":3999},{"type":21,"tag":322,"props":27108,"children":27109},{"class":324,"line":5087},[27110,27114],{"type":21,"tag":322,"props":27111,"children":27112},{"style":3638},[27113],{"type":26,"value":5023},{"type":21,"tag":322,"props":27115,"children":27116},{"style":329},[27117],{"type":26,"value":27118}," qml.expval(grouped_H)\n",{"type":21,"tag":322,"props":27120,"children":27121},{"class":324,"line":5095},[27122],{"type":21,"tag":322,"props":27123,"children":27124},{"emptyLinePlaceholder":3662},[27125],{"type":26,"value":3665},{"type":21,"tag":322,"props":27127,"children":27128},{"class":324,"line":5126},[27129],{"type":21,"tag":322,"props":27130,"children":27131},{"style":3888},[27132],{"type":26,"value":27133},"# 2. Use shot-adaptive optimizer\n",{"type":21,"tag":322,"props":27135,"children":27136},{"class":324,"line":5171},[27137,27141,27145,27149,27153,27157,27162],{"type":21,"tag":322,"props":27138,"children":27139},{"style":329},[27140],{"type":26,"value":15389},{"type":21,"tag":322,"props":27142,"children":27143},{"style":3638},[27144],{"type":26,"value":3678},{"type":21,"tag":322,"props":27146,"children":27147},{"style":329},[27148],{"type":26,"value":25640},{"type":21,"tag":322,"props":27150,"children":27151},{"style":3703},[27152],{"type":26,"value":25645},{"type":21,"tag":322,"props":27154,"children":27155},{"style":3638},[27156],{"type":26,"value":3678},{"type":21,"tag":322,"props":27158,"children":27159},{"style":348},[27160],{"type":26,"value":27161},"50",{"type":21,"tag":322,"props":27163,"children":27164},{"style":329},[27165],{"type":26,"value":3738},{"type":21,"tag":322,"props":27167,"children":27168},{"class":324,"line":5210},[27169],{"type":21,"tag":322,"props":27170,"children":27171},{"emptyLinePlaceholder":3662},[27172],{"type":26,"value":3665},{"type":21,"tag":322,"props":27174,"children":27175},{"class":324,"line":5228},[27176],{"type":21,"tag":322,"props":27177,"children":27178},{"style":3888},[27179],{"type":26,"value":27180},"# 3. Warm start\n",{"type":21,"tag":322,"props":27182,"children":27183},{"class":324,"line":5302},[27184,27188,27192],{"type":21,"tag":322,"props":27185,"children":27186},{"style":329},[27187],{"type":26,"value":25672},{"type":21,"tag":322,"props":27189,"children":27190},{"style":3638},[27191],{"type":26,"value":3678},{"type":21,"tag":322,"props":27193,"children":27194},{"style":329},[27195],{"type":26,"value":27196}," warm_start_params(H)\n",{"type":21,"tag":322,"props":27198,"children":27199},{"class":324,"line":5330},[27200],{"type":21,"tag":322,"props":27201,"children":27202},{"emptyLinePlaceholder":3662},[27203],{"type":26,"value":3665},{"type":21,"tag":322,"props":27205,"children":27206},{"class":324,"line":5359},[27207],{"type":21,"tag":322,"props":27208,"children":27209},{"style":3888},[27210],{"type":26,"value":27211},"# 4. Run with early stopping\n",{"type":21,"tag":322,"props":27213,"children":27214},{"class":324,"line":5368},[27215,27219,27223,27227,27231,27235,27239],{"type":21,"tag":322,"props":27216,"children":27217},{"style":3638},[27218],{"type":26,"value":3931},{"type":21,"tag":322,"props":27220,"children":27221},{"style":329},[27222],{"type":26,"value":15451},{"type":21,"tag":322,"props":27224,"children":27225},{"style":3638},[27226],{"type":26,"value":3941},{"type":21,"tag":322,"props":27228,"children":27229},{"style":348},[27230],{"type":26,"value":3946},{"type":21,"tag":322,"props":27232,"children":27233},{"style":329},[27234],{"type":26,"value":3758},{"type":21,"tag":322,"props":27236,"children":27237},{"style":348},[27238],{"type":26,"value":26150},{"type":21,"tag":322,"props":27240,"children":27241},{"style":329},[27242],{"type":26,"value":4851},{"type":21,"tag":322,"props":27244,"children":27245},{"class":324,"line":5396},[27246,27251,27255],{"type":21,"tag":322,"props":27247,"children":27248},{"style":329},[27249],{"type":26,"value":27250},"    params, energy, shots ",{"type":21,"tag":322,"props":27252,"children":27253},{"style":3638},[27254],{"type":26,"value":3678},{"type":21,"tag":322,"props":27256,"children":27257},{"style":329},[27258],{"type":26,"value":27259}," opt.step_and_cost(ansatz, params)\n",{"type":21,"tag":322,"props":27261,"children":27262},{"class":324,"line":5427},[27263,27267],{"type":21,"tag":322,"props":27264,"children":27265},{"style":3638},[27266],{"type":26,"value":12384},{"type":21,"tag":322,"props":27268,"children":27269},{"style":329},[27270],{"type":26,"value":27271}," check_convergence(energy, shots):\n",{"type":21,"tag":322,"props":27273,"children":27274},{"class":324,"line":5449},[27275],{"type":21,"tag":322,"props":27276,"children":27277},{"style":3638},[27278],{"type":26,"value":26549},{"type":21,"tag":322,"props":27280,"children":27281},{"class":324,"line":5495},[27282],{"type":21,"tag":322,"props":27283,"children":27284},{"emptyLinePlaceholder":3662},[27285],{"type":26,"value":3665},{"type":21,"tag":322,"props":27287,"children":27288},{"class":324,"line":5508},[27289,27293,27297,27301,27306,27310,27315,27319,27323,27327],{"type":21,"tag":322,"props":27290,"children":27291},{"style":348},[27292],{"type":26,"value":3753},{"type":21,"tag":322,"props":27294,"children":27295},{"style":329},[27296],{"type":26,"value":3758},{"type":21,"tag":322,"props":27298,"children":27299},{"style":3638},[27300],{"type":26,"value":3763},{"type":21,"tag":322,"props":27302,"children":27303},{"style":359},[27304],{"type":26,"value":27305},"\"Ground state energy: ",{"type":21,"tag":322,"props":27307,"children":27308},{"style":348},[27309],{"type":26,"value":3773},{"type":21,"tag":322,"props":27311,"children":27312},{"style":329},[27313],{"type":26,"value":27314},"energy",{"type":21,"tag":322,"props":27316,"children":27317},{"style":3638},[27318],{"type":26,"value":4085},{"type":21,"tag":322,"props":27320,"children":27321},{"style":348},[27322],{"type":26,"value":3783},{"type":21,"tag":322,"props":27324,"children":27325},{"style":359},[27326],{"type":26,"value":19868},{"type":21,"tag":322,"props":27328,"children":27329},{"style":329},[27330],{"type":26,"value":3738},{"type":21,"tag":59,"props":27332,"children":27334},{"id":27333},"クイックリファレンス-手法別のショット予算",[27335],{"type":26,"value":27336},"クイックリファレンス: 手法別のショット予算",{"type":21,"tag":4503,"props":27338,"children":27339},{},[27340,27361],{"type":21,"tag":4507,"props":27341,"children":27342},{},[27343],{"type":21,"tag":4511,"props":27344,"children":27345},{},[27346,27351,27356],{"type":21,"tag":4515,"props":27347,"children":27348},{},[27349],{"type":26,"value":27350},"手法",{"type":21,"tag":4515,"props":27352,"children":27353},{},[27354],{"type":26,"value":27355},"ショット数\u002Fステップ",{"type":21,"tag":4515,"props":27357,"children":27358},{},[27359],{"type":26,"value":27360},"最適な場面",{"type":21,"tag":4529,"props":27362,"children":27363},{},[27364,27386,27408,27430,27448],{"type":21,"tag":4511,"props":27365,"children":27366},{},[27367,27372,27381],{"type":21,"tag":4536,"props":27368,"children":27369},{},[27370],{"type":26,"value":27371},"COBYLA + 固定ショット",{"type":21,"tag":4536,"props":27373,"children":27374},{},[27375],{"type":21,"tag":79,"props":27376,"children":27378},{"className":27377},[],[27379],{"type":26,"value":27380},"n_terms × shots",{"type":21,"tag":4536,"props":27382,"children":27383},{},[27384],{"type":26,"value":27385},"パラメータ数が少ない場合",{"type":21,"tag":4511,"props":27387,"children":27388},{},[27389,27394,27403],{"type":21,"tag":4536,"props":27390,"children":27391},{},[27392],{"type":26,"value":27393},"パラメータシフト + Adam",{"type":21,"tag":4536,"props":27395,"children":27396},{},[27397],{"type":21,"tag":79,"props":27398,"children":27400},{"className":27399},[],[27401],{"type":26,"value":27402},"2n_params × shots",{"type":21,"tag":4536,"props":27404,"children":27405},{},[27406],{"type":26,"value":27407},"微分可能な回路",{"type":21,"tag":4511,"props":27409,"children":27410},{},[27411,27416,27425],{"type":21,"tag":4536,"props":27412,"children":27413},{},[27414],{"type":26,"value":27415},"SPSA",{"type":21,"tag":4536,"props":27417,"children":27418},{},[27419],{"type":21,"tag":79,"props":27420,"children":27422},{"className":27421},[],[27423],{"type":26,"value":27424},"2 × shots",{"type":21,"tag":4536,"props":27426,"children":27427},{},[27428],{"type":26,"value":27429},"パラメータ数が多い場合",{"type":21,"tag":4511,"props":27431,"children":27432},{},[27433,27438,27443],{"type":21,"tag":4536,"props":27434,"children":27435},{},[27436],{"type":26,"value":27437},"ShotAdaptiveOptimizer",{"type":21,"tag":4536,"props":27439,"children":27440},{},[27441],{"type":26,"value":27442},"適応的",{"type":21,"tag":4536,"props":27444,"children":27445},{},[27446],{"type":26,"value":27447},"一般的なVQE",{"type":21,"tag":4511,"props":27449,"children":27450},{},[27451,27456,27461],{"type":21,"tag":4536,"props":27452,"children":27453},{},[27454],{"type":26,"value":27455},"グループ化されたパウリ項",{"type":21,"tag":4536,"props":27457,"children":27458},{},[27459],{"type":26,"value":27460},"÷5〜15×",{"type":21,"tag":4536,"props":27462,"children":27463},{},[27464],{"type":26,"value":27465},"常に最初に適用",{"type":21,"tag":22,"props":27467,"children":27468},{},[27469],{"type":26,"value":27470},"まず測定のグループ化を適用してください。これは単一で最大の効果を持ち、オプティマイザや回路に変更を加える必要がありません。",{"type":21,"tag":22,"props":27472,"children":27473},{},[27474,27479,27480,27485,27486,27491,27492],{"type":21,"tag":71,"props":27475,"children":27476},{},[27477],{"type":26,"value":27478},"関連:",{"type":26,"value":1901},{"type":21,"tag":29,"props":27481,"children":27482},{"href":893},[27483],{"type":26,"value":27484},"VQE with PennyLane",{"type":26,"value":7549},{"type":21,"tag":29,"props":27487,"children":27488},{"href":806},[27489],{"type":26,"value":27490},"QPU access guide",{"type":26,"value":7549},{"type":21,"tag":29,"props":27493,"children":27494},{"href":19441},[27495],{"type":26,"value":19444},{"type":21,"tag":703,"props":27497,"children":27498},{},[27499],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":27501},[27502,27503,27504,27505,27506,27507,27508,27509,27510],{"id":24987,"depth":335,"text":24987},{"id":25130,"depth":335,"text":25133},{"id":25458,"depth":335,"text":25461},{"id":25811,"depth":335,"text":25814},{"id":26055,"depth":335,"text":26058},{"id":26232,"depth":335,"text":26235},{"id":26581,"depth":335,"text":26584},{"id":26765,"depth":335,"text":26768},{"id":27333,"depth":335,"text":27336},"content:ja:blog:reduce-shot-count.md","ja\u002Fblog\u002Freduce-shot-count.md","ja\u002Fblog\u002Freduce-shot-count",{"_path":27515,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":27516,"description":27517,"date":27518,"updatedDate":10,"author":11,"tags":27519,"readingTime":496,"body":27521,"_type":715,"_id":29324,"_source":717,"_file":29325,"_stem":29326,"_extension":720},"\u002Fja\u002Fblog\u002Fibm-quantum-free-tier-2026","IBM Quantum 無料プラン: 2026年に利用できるすべて","IBM Quantum の無料 Open Plan に含まれる内容、キュー待ち時間の仕組み、そして無料 QPU アクセスを最大限に活用する方法を詳しく解説します。","2026-02-20",[27520,730,1962],"IBM",{"type":18,"children":27522,"toc":29310},[27523,27528,27533,27545,27602,27607,27613,27641,27662,27682,27688,27693,27798,27803,28045,28051,28057,28067,28073,28100,28106,28111,28222,28235,28241,28885,28890,28895,28921,28929,28960,28965,28970,29049,29054,29060,29065,29294,29306],{"type":21,"tag":22,"props":27524,"children":27525},{},[27526],{"type":26,"value":27527},"IBM Quantum は、量子コンピューティング業界で最もアクセスしやすい無料プランを提供しています。無料の IBM ID があれば、最大 127 量子ビットの実際の量子プロセッサにアクセスできます。費用も、クレジットカードも、申請も不要です。それが実際に何を意味するのか、正確に見ていきましょう。",{"type":21,"tag":59,"props":27529,"children":27531},{"id":27530},"無料プランに含まれるもの",[27532],{"type":26,"value":27530},{"type":21,"tag":22,"props":27534,"children":27535},{},[27536,27538,27543],{"type":26,"value":27537},"IBM Quantum の ",{"type":21,"tag":71,"props":27539,"children":27540},{},[27541],{"type":26,"value":27542},"Open Plan",{"type":26,"value":27544}," では以下が利用できます:",{"type":21,"tag":3556,"props":27546,"children":27547},{},[27548,27558,27570,27580,27590],{"type":21,"tag":3560,"props":27549,"children":27550},{},[27551,27553],{"type":26,"value":27552},"一般公開されているすべての量子システムへの",{"type":21,"tag":71,"props":27554,"children":27555},{},[27556],{"type":26,"value":27557},"無制限アクセス",{"type":21,"tag":3560,"props":27559,"children":27560},{},[27561,27563,27568],{"type":26,"value":27562},"月あたり ",{"type":21,"tag":71,"props":27564,"children":27565},{},[27566],{"type":26,"value":27567},"10分の量子時間",{"type":26,"value":27569},"(回路が QPU 上で実際に実行されている時間)",{"type":21,"tag":3560,"props":27571,"children":27572},{},[27573,27578],{"type":21,"tag":71,"props":27574,"children":27575},{},[27576],{"type":26,"value":27577},"IBM シミュレータへのアクセス",{"type":26,"value":27579},"(時間制限なし。statevector、QASM、matrix product state)",{"type":21,"tag":3560,"props":27581,"children":27582},{},[27583,27588],{"type":21,"tag":71,"props":27584,"children":27585},{},[27586],{"type":26,"value":27587},"Qiskit Runtime",{"type":26,"value":27589},": 回路を実行するための最新の IBM 実行環境",{"type":21,"tag":3560,"props":27591,"children":27592},{},[27593,27595,27600],{"type":26,"value":27594},"チュートリアルやコースを備えた ",{"type":21,"tag":71,"props":27596,"children":27597},{},[27598],{"type":26,"value":27599},"IBM Quantum Learning",{"type":26,"value":27601}," プラットフォーム",{"type":21,"tag":22,"props":27603,"children":27604},{},[27605],{"type":26,"value":27606},"月10分の制限は厳しく聞こえますが、実際にはほとんどの研究用回路は数ミリ秒の実 QPU 時間で完了します。量子アルゴリズムを試している個人開発者がこの制限に達することはめったにありません。",{"type":21,"tag":59,"props":27608,"children":27610},{"id":27609},"_2026年8月アップデートheron-r2とボーナス計算時間",[27611],{"type":26,"value":27612},"2026年8月アップデート:Heron r2とボーナス計算時間",{"type":21,"tag":22,"props":27614,"children":27615},{},[27616,27618,27624,27626,27632,27634,27639],{"type":26,"value":27617},"IBMは2026年3月、より高性能なHeron r2プロセッサ(",{"type":21,"tag":79,"props":27619,"children":27621},{"className":27620},[],[27622],{"type":26,"value":27623},"ibm_kingston",{"type":26,"value":27625},")をOpen Planに追加し、無料枠ユーザーは従来のHeron r1システムより明らかに優れた2量子ビットゲート忠実度にアクセスできるようになりました。しばらく実行していない場合は、",{"type":21,"tag":79,"props":27627,"children":27629},{"className":27628},[],[27630],{"type":26,"value":27631},"service.backends()",{"type":26,"value":27633},"のリストに",{"type":21,"tag":79,"props":27635,"children":27637},{"className":27636},[],[27638],{"type":26,"value":27623},{"type":26,"value":27640},"が表示されているか確認する価値があります。",{"type":21,"tag":22,"props":27642,"children":27643},{},[27644,27646,27651,27653,27660],{"type":26,"value":27645},"IBMはまた、アクティブユーザー向けに1回限りのプロモーションを実施しました。12か月間のいずれかの期間に少なくとも20分のQPU時間を記録したユーザーは、以降12か月間で",{"type":21,"tag":71,"props":27647,"children":27648},{},[27649],{"type":26,"value":27650},"180分",{"type":26,"value":27652},"のボーナス(標準の月間割り当ての18倍)を選択できました。この特典は特定の登録期間に限定されたものであり、Open Planの月10分という基本枠を恒久的に変更するものではありません。したがって、まだ有効だと思い込まず、",{"type":21,"tag":29,"props":27654,"children":27657},{"href":27655,"rel":27656},"https:\u002F\u002Fquantum.cloud.ibm.com\u002Fdocs\u002Fen\u002Fguides\u002Fplans-overview",[44],[27658],{"type":26,"value":27659},"Open Planページ",{"type":26,"value":27661},"でご自身のアカウントに現在何が適用されているか確認してください。",{"type":21,"tag":22,"props":27663,"children":27664},{},[27665,27667,27673,27675,27680],{"type":26,"value":27666},"システムの構成は逆方向にも変化します。IBMは2026年4月初旬に",{"type":21,"tag":79,"props":27668,"children":27670},{"className":27669},[],[27671],{"type":26,"value":27672},"ibm_torino",{"type":26,"value":27674},"を廃止しました。これはOpen Planの運用における通常の一部です。新しいハードウェアが登場すればシステムが追加され、既存のシステムは耐用年数を終えると廃止されます。だからこそ、下記の",{"type":21,"tag":79,"props":27676,"children":27678},{"className":27677},[],[27679],{"type":26,"value":27631},{"type":26,"value":27681},"のスニペットは、どんな静的な表よりも重要です。この記事にある一覧は現在の状態ではなく、あくまである時点のスナップショットとして扱ってください。",{"type":21,"tag":59,"props":27683,"children":27685},{"id":27684},"利用可能な無料-qpu",[27686],{"type":26,"value":27687},"利用可能な無料 QPU",{"type":21,"tag":22,"props":27689,"children":27690},{},[27691],{"type":26,"value":27692},"2026年半ば時点で、IBMのOpen Planは次のようなシステムへのアクセスを含みます:",{"type":21,"tag":4503,"props":27694,"children":27695},{},[27696,27721],{"type":21,"tag":4507,"props":27697,"children":27698},{},[27699],{"type":21,"tag":4511,"props":27700,"children":27701},{},[27702,27707,27711,27716],{"type":21,"tag":4515,"props":27703,"children":27704},{},[27705],{"type":26,"value":27706},"システム",{"type":21,"tag":4515,"props":27708,"children":27709},{},[27710],{"type":26,"value":3567},{"type":21,"tag":4515,"props":27712,"children":27713},{},[27714],{"type":26,"value":27715},"技術",{"type":21,"tag":4515,"props":27717,"children":27718},{},[27719],{"type":26,"value":27720},"一般的なキュー",{"type":21,"tag":4529,"props":27722,"children":27723},{},[27724,27753,27776],{"type":21,"tag":4511,"props":27725,"children":27726},{},[27727,27738,27743,27748],{"type":21,"tag":4536,"props":27728,"children":27729},{},[27730,27732,27737],{"type":26,"value":27731},"IBM Heron r2(",{"type":21,"tag":79,"props":27733,"children":27735},{"className":27734},[],[27736],{"type":26,"value":27623},{"type":26,"value":17711},{"type":21,"tag":4536,"props":27739,"children":27740},{},[27741],{"type":26,"value":27742},"156",{"type":21,"tag":4536,"props":27744,"children":27745},{},[27746],{"type":26,"value":27747},"Heronプロセッサ、高忠実度",{"type":21,"tag":4536,"props":27749,"children":27750},{},[27751],{"type":26,"value":27752},"30〜120分",{"type":21,"tag":4511,"props":27754,"children":27755},{},[27756,27761,27766,27771],{"type":21,"tag":4536,"props":27757,"children":27758},{},[27759],{"type":26,"value":27760},"IBM Eagle",{"type":21,"tag":4536,"props":27762,"children":27763},{},[27764],{"type":26,"value":27765},"127",{"type":21,"tag":4536,"props":27767,"children":27768},{},[27769],{"type":26,"value":27770},"超伝導",{"type":21,"tag":4536,"props":27772,"children":27773},{},[27774],{"type":26,"value":27775},"10〜60分",{"type":21,"tag":4511,"props":27777,"children":27778},{},[27779,27784,27789,27794],{"type":21,"tag":4536,"props":27780,"children":27781},{},[27782],{"type":26,"value":27783},"IBM Heron r1",{"type":21,"tag":4536,"props":27785,"children":27786},{},[27787],{"type":26,"value":27788},"133",{"type":21,"tag":4536,"props":27790,"children":27791},{},[27792],{"type":26,"value":27793},"Heron プロセッサ",{"type":21,"tag":4536,"props":27795,"children":27796},{},[27797],{"type":26,"value":27752},{"type":21,"tag":22,"props":27799,"children":27800},{},[27801],{"type":26,"value":27802},"IBMは定期的にシステムを廃止・追加しています。この表はあくまでスナップショットであり、保証ではありません。現在のリストを確認するには Qiskit Runtime サービスを使用してください:",{"type":21,"tag":312,"props":27804,"children":27806},{"className":3626,"code":27805,"language":3628,"meta":7,"style":7},"from qiskit_ibm_runtime import QiskitRuntimeService\n\nservice = QiskitRuntimeService(channel=\"ibm_quantum\")\nbackends = service.backends(\n    operational=True,\n    simulator=False,\n    min_num_qubits=1\n)\nfor b in backends:\n    print(f\"{b.name}: {b.num_qubits} qubits, status: {b.status().status_msg}\")\n",[27807],{"type":21,"tag":79,"props":27808,"children":27809},{"__ignoreMap":7},[27810,27829,27836,27869,27886,27906,27926,27943,27950,27970],{"type":21,"tag":322,"props":27811,"children":27812},{"class":324,"line":325},[27813,27817,27821,27825],{"type":21,"tag":322,"props":27814,"children":27815},{"style":3638},[27816],{"type":26,"value":3641},{"type":21,"tag":322,"props":27818,"children":27819},{"style":329},[27820],{"type":26,"value":3646},{"type":21,"tag":322,"props":27822,"children":27823},{"style":3638},[27824],{"type":26,"value":3651},{"type":21,"tag":322,"props":27826,"children":27827},{"style":329},[27828],{"type":26,"value":3656},{"type":21,"tag":322,"props":27830,"children":27831},{"class":324,"line":335},[27832],{"type":21,"tag":322,"props":27833,"children":27834},{"emptyLinePlaceholder":3662},[27835],{"type":26,"value":3665},{"type":21,"tag":322,"props":27837,"children":27838},{"class":324,"line":344},[27839,27843,27847,27852,27857,27861,27865],{"type":21,"tag":322,"props":27840,"children":27841},{"style":329},[27842],{"type":26,"value":3673},{"type":21,"tag":322,"props":27844,"children":27845},{"style":3638},[27846],{"type":26,"value":3678},{"type":21,"tag":322,"props":27848,"children":27849},{"style":329},[27850],{"type":26,"value":27851}," QiskitRuntimeService(",{"type":21,"tag":322,"props":27853,"children":27854},{"style":3703},[27855],{"type":26,"value":27856},"channel",{"type":21,"tag":322,"props":27858,"children":27859},{"style":3638},[27860],{"type":26,"value":3678},{"type":21,"tag":322,"props":27862,"children":27863},{"style":359},[27864],{"type":26,"value":21163},{"type":21,"tag":322,"props":27866,"children":27867},{"style":329},[27868],{"type":26,"value":3738},{"type":21,"tag":322,"props":27870,"children":27871},{"class":324,"line":407},[27872,27877,27881],{"type":21,"tag":322,"props":27873,"children":27874},{"style":329},[27875],{"type":26,"value":27876},"backends ",{"type":21,"tag":322,"props":27878,"children":27879},{"style":3638},[27880],{"type":26,"value":3678},{"type":21,"tag":322,"props":27882,"children":27883},{"style":329},[27884],{"type":26,"value":27885}," service.backends(\n",{"type":21,"tag":322,"props":27887,"children":27888},{"class":324,"line":431},[27889,27894,27898,27902],{"type":21,"tag":322,"props":27890,"children":27891},{"style":3703},[27892],{"type":26,"value":27893},"    operational",{"type":21,"tag":322,"props":27895,"children":27896},{"style":3638},[27897],{"type":26,"value":3678},{"type":21,"tag":322,"props":27899,"children":27900},{"style":348},[27901],{"type":26,"value":3715},{"type":21,"tag":322,"props":27903,"children":27904},{"style":329},[27905],{"type":26,"value":428},{"type":21,"tag":322,"props":27907,"children":27908},{"class":324,"line":453},[27909,27914,27918,27922],{"type":21,"tag":322,"props":27910,"children":27911},{"style":3703},[27912],{"type":26,"value":27913},"    simulator",{"type":21,"tag":322,"props":27915,"children":27916},{"style":3638},[27917],{"type":26,"value":3678},{"type":21,"tag":322,"props":27919,"children":27920},{"style":348},[27921],{"type":26,"value":3733},{"type":21,"tag":322,"props":27923,"children":27924},{"style":329},[27925],{"type":26,"value":428},{"type":21,"tag":322,"props":27927,"children":27928},{"class":324,"line":474},[27929,27934,27938],{"type":21,"tag":322,"props":27930,"children":27931},{"style":3703},[27932],{"type":26,"value":27933},"    min_num_qubits",{"type":21,"tag":322,"props":27935,"children":27936},{"style":3638},[27937],{"type":26,"value":3678},{"type":21,"tag":322,"props":27939,"children":27940},{"style":348},[27941],{"type":26,"value":27942},"1\n",{"type":21,"tag":322,"props":27944,"children":27945},{"class":324,"line":496},[27946],{"type":21,"tag":322,"props":27947,"children":27948},{"style":329},[27949],{"type":26,"value":3738},{"type":21,"tag":322,"props":27951,"children":27952},{"class":324,"line":16},[27953,27957,27961,27965],{"type":21,"tag":322,"props":27954,"children":27955},{"style":3638},[27956],{"type":26,"value":3931},{"type":21,"tag":322,"props":27958,"children":27959},{"style":329},[27960],{"type":26,"value":5341},{"type":21,"tag":322,"props":27962,"children":27963},{"style":3638},[27964],{"type":26,"value":3941},{"type":21,"tag":322,"props":27966,"children":27967},{"style":329},[27968],{"type":26,"value":27969}," backends:\n",{"type":21,"tag":322,"props":27971,"children":27972},{"class":324,"line":3884},[27973,27977,27981,27985,27989,27993,27998,28002,28006,28010,28015,28019,28024,28028,28033,28037,28041],{"type":21,"tag":322,"props":27974,"children":27975},{"style":348},[27976],{"type":26,"value":17663},{"type":21,"tag":322,"props":27978,"children":27979},{"style":329},[27980],{"type":26,"value":3758},{"type":21,"tag":322,"props":27982,"children":27983},{"style":3638},[27984],{"type":26,"value":3763},{"type":21,"tag":322,"props":27986,"children":27987},{"style":359},[27988],{"type":26,"value":3788},{"type":21,"tag":322,"props":27990,"children":27991},{"style":348},[27992],{"type":26,"value":3773},{"type":21,"tag":322,"props":27994,"children":27995},{"style":329},[27996],{"type":26,"value":27997},"b.name",{"type":21,"tag":322,"props":27999,"children":28000},{"style":348},[28001],{"type":26,"value":3783},{"type":21,"tag":322,"props":28003,"children":28004},{"style":359},[28005],{"type":26,"value":418},{"type":21,"tag":322,"props":28007,"children":28008},{"style":348},[28009],{"type":26,"value":3773},{"type":21,"tag":322,"props":28011,"children":28012},{"style":329},[28013],{"type":26,"value":28014},"b.num_qubits",{"type":21,"tag":322,"props":28016,"children":28017},{"style":348},[28018],{"type":26,"value":3783},{"type":21,"tag":322,"props":28020,"children":28021},{"style":359},[28022],{"type":26,"value":28023}," qubits, status: ",{"type":21,"tag":322,"props":28025,"children":28026},{"style":348},[28027],{"type":26,"value":3773},{"type":21,"tag":322,"props":28029,"children":28030},{"style":329},[28031],{"type":26,"value":28032},"b.status().status_msg",{"type":21,"tag":322,"props":28034,"children":28035},{"style":348},[28036],{"type":26,"value":3783},{"type":21,"tag":322,"props":28038,"children":28039},{"style":359},[28040],{"type":26,"value":3788},{"type":21,"tag":322,"props":28042,"children":28043},{"style":329},[28044],{"type":26,"value":3738},{"type":21,"tag":59,"props":28046,"children":28048},{"id":28047},"_10分でのセットアップ",[28049],{"type":26,"value":28050},"10分でのセットアップ",{"type":21,"tag":17800,"props":28052,"children":28054},{"id":28053},"ステップ1-無料アカウントの作成",[28055],{"type":26,"value":28056},"ステップ1: 無料アカウントの作成",{"type":21,"tag":22,"props":28058,"children":28059},{},[28060,28065],{"type":21,"tag":29,"props":28061,"children":28063},{"href":21089,"rel":28062},[44],[28064],{"type":26,"value":21093},{"type":26,"value":28066}," にアクセスし、無料の IBM ID を作成します。支払い情報は不要です。",{"type":21,"tag":17800,"props":28068,"children":28070},{"id":28069},"ステップ2-sdk-のインストール",[28071],{"type":26,"value":28072},"ステップ2: SDK のインストール",{"type":21,"tag":312,"props":28074,"children":28076},{"className":10181,"code":28075,"language":10183,"meta":7,"style":7},"pip install qiskit qiskit-ibm-runtime\n",[28077],{"type":21,"tag":79,"props":28078,"children":28079},{"__ignoreMap":7},[28080],{"type":21,"tag":322,"props":28081,"children":28082},{"class":324,"line":325},[28083,28087,28091,28095],{"type":21,"tag":322,"props":28084,"children":28085},{"style":4801},[28086],{"type":26,"value":12232},{"type":21,"tag":322,"props":28088,"children":28089},{"style":359},[28090],{"type":26,"value":12237},{"type":21,"tag":322,"props":28092,"children":28093},{"style":359},[28094],{"type":26,"value":12242},{"type":21,"tag":322,"props":28096,"children":28097},{"style":359},[28098],{"type":26,"value":28099}," qiskit-ibm-runtime\n",{"type":21,"tag":17800,"props":28101,"children":28103},{"id":28102},"ステップ3-認証情報の保存",[28104],{"type":26,"value":28105},"ステップ3: 認証情報の保存",{"type":21,"tag":22,"props":28107,"children":28108},{},[28109],{"type":26,"value":28110},"IBM Quantum ダッシュボードにログインした後、Account ページから API トークンをコピーします。",{"type":21,"tag":312,"props":28112,"children":28114},{"className":3626,"code":28113,"language":3628,"meta":7,"style":7},"from qiskit_ibm_runtime import QiskitRuntimeService\n\n# Run this once to save your token locally\nQiskitRuntimeService.save_account(\n    channel=\"ibm_quantum\",\n    token=\"YOUR_API_TOKEN_HERE\",\n    overwrite=True\n)\n",[28115],{"type":21,"tag":79,"props":28116,"children":28117},{"__ignoreMap":7},[28118,28137,28144,28152,28159,28178,28198,28215],{"type":21,"tag":322,"props":28119,"children":28120},{"class":324,"line":325},[28121,28125,28129,28133],{"type":21,"tag":322,"props":28122,"children":28123},{"style":3638},[28124],{"type":26,"value":3641},{"type":21,"tag":322,"props":28126,"children":28127},{"style":329},[28128],{"type":26,"value":3646},{"type":21,"tag":322,"props":28130,"children":28131},{"style":3638},[28132],{"type":26,"value":3651},{"type":21,"tag":322,"props":28134,"children":28135},{"style":329},[28136],{"type":26,"value":3656},{"type":21,"tag":322,"props":28138,"children":28139},{"class":324,"line":335},[28140],{"type":21,"tag":322,"props":28141,"children":28142},{"emptyLinePlaceholder":3662},[28143],{"type":26,"value":3665},{"type":21,"tag":322,"props":28145,"children":28146},{"class":324,"line":344},[28147],{"type":21,"tag":322,"props":28148,"children":28149},{"style":3888},[28150],{"type":26,"value":28151},"# Run this once to save your token locally\n",{"type":21,"tag":322,"props":28153,"children":28154},{"class":324,"line":407},[28155],{"type":21,"tag":322,"props":28156,"children":28157},{"style":329},[28158],{"type":26,"value":21146},{"type":21,"tag":322,"props":28160,"children":28161},{"class":324,"line":431},[28162,28166,28170,28174],{"type":21,"tag":322,"props":28163,"children":28164},{"style":3703},[28165],{"type":26,"value":21154},{"type":21,"tag":322,"props":28167,"children":28168},{"style":3638},[28169],{"type":26,"value":3678},{"type":21,"tag":322,"props":28171,"children":28172},{"style":359},[28173],{"type":26,"value":21163},{"type":21,"tag":322,"props":28175,"children":28176},{"style":329},[28177],{"type":26,"value":428},{"type":21,"tag":322,"props":28179,"children":28180},{"class":324,"line":453},[28181,28185,28189,28194],{"type":21,"tag":322,"props":28182,"children":28183},{"style":3703},[28184],{"type":26,"value":21175},{"type":21,"tag":322,"props":28186,"children":28187},{"style":3638},[28188],{"type":26,"value":3678},{"type":21,"tag":322,"props":28190,"children":28191},{"style":359},[28192],{"type":26,"value":28193},"\"YOUR_API_TOKEN_HERE\"",{"type":21,"tag":322,"props":28195,"children":28196},{"style":329},[28197],{"type":26,"value":428},{"type":21,"tag":322,"props":28199,"children":28200},{"class":324,"line":474},[28201,28206,28210],{"type":21,"tag":322,"props":28202,"children":28203},{"style":3703},[28204],{"type":26,"value":28205},"    overwrite",{"type":21,"tag":322,"props":28207,"children":28208},{"style":3638},[28209],{"type":26,"value":3678},{"type":21,"tag":322,"props":28211,"children":28212},{"style":348},[28213],{"type":26,"value":28214},"True\n",{"type":21,"tag":322,"props":28216,"children":28217},{"class":324,"line":496},[28218],{"type":21,"tag":322,"props":28219,"children":28220},{"style":329},[28221],{"type":26,"value":3738},{"type":21,"tag":22,"props":28223,"children":28224},{},[28225,28227,28233],{"type":26,"value":28226},"認証情報は ",{"type":21,"tag":79,"props":28228,"children":28230},{"className":28229},[],[28231],{"type":26,"value":28232},"~\u002F.qiskit\u002Fqiskit-ibm.json",{"type":26,"value":28234}," に保存されます。次回以降はこの操作は不要です。",{"type":21,"tag":17800,"props":28236,"children":28238},{"id":28237},"ステップ4-最初のジョブを投入する",[28239],{"type":26,"value":28240},"ステップ4: 最初のジョブを投入する",{"type":21,"tag":312,"props":28242,"children":28244},{"className":3626,"code":28243,"language":3628,"meta":7,"style":7},"from qiskit import QuantumCircuit\nfrom qiskit.transpiler.preset_passmanagers import generate_preset_pass_manager\nfrom qiskit_ibm_runtime import QiskitRuntimeService, SamplerV2 as Sampler\n\n# Build a Bell state\nqc = QuantumCircuit(2, 2)\nqc.h(0)\nqc.cx(0, 1)\nqc.measure([0, 1], [0, 1])\n\n# Connect and pick the least-busy free QPU\nservice = QiskitRuntimeService(channel=\"ibm_quantum\")\nbackend = service.least_busy(\n    operational=True,\n    simulator=False,\n    min_num_qubits=2\n)\nprint(f\"Submitting to: {backend.name}\")\n\n# Transpile for the target device\npm = generate_preset_pass_manager(backend=backend, optimization_level=1)\nisa_qc = pm.run(qc)\n\n# Submit\nsampler = Sampler(mode=backend)\njob = sampler.run([isa_qc], shots=1024)\n\nprint(f\"Job ID: {job.job_id()}\")\nprint(f\"Status: {job.status()}\")\nresult = job.result()\nprint(result[0].data.c.get_counts())\n",[28245],{"type":21,"tag":79,"props":28246,"children":28247},{"__ignoreMap":7},[28248,28267,28288,28317,28324,28332,28363,28378,28401,28440,28447,28455,28486,28502,28521,28540,28556,28563,28603,28610,28618,28663,28680,28687,28695,28726,28759,28766,28807,28848,28864],{"type":21,"tag":322,"props":28249,"children":28250},{"class":324,"line":325},[28251,28255,28259,28263],{"type":21,"tag":322,"props":28252,"children":28253},{"style":3638},[28254],{"type":26,"value":3641},{"type":21,"tag":322,"props":28256,"children":28257},{"style":329},[28258],{"type":26,"value":4732},{"type":21,"tag":322,"props":28260,"children":28261},{"style":3638},[28262],{"type":26,"value":3651},{"type":21,"tag":322,"props":28264,"children":28265},{"style":329},[28266],{"type":26,"value":20739},{"type":21,"tag":322,"props":28268,"children":28269},{"class":324,"line":335},[28270,28274,28279,28283],{"type":21,"tag":322,"props":28271,"children":28272},{"style":3638},[28273],{"type":26,"value":3641},{"type":21,"tag":322,"props":28275,"children":28276},{"style":329},[28277],{"type":26,"value":28278}," qiskit.transpiler.preset_passmanagers ",{"type":21,"tag":322,"props":28280,"children":28281},{"style":3638},[28282],{"type":26,"value":3651},{"type":21,"tag":322,"props":28284,"children":28285},{"style":329},[28286],{"type":26,"value":28287}," generate_preset_pass_manager\n",{"type":21,"tag":322,"props":28289,"children":28290},{"class":324,"line":344},[28291,28295,28299,28303,28308,28312],{"type":21,"tag":322,"props":28292,"children":28293},{"style":3638},[28294],{"type":26,"value":3641},{"type":21,"tag":322,"props":28296,"children":28297},{"style":329},[28298],{"type":26,"value":3646},{"type":21,"tag":322,"props":28300,"children":28301},{"style":3638},[28302],{"type":26,"value":3651},{"type":21,"tag":322,"props":28304,"children":28305},{"style":329},[28306],{"type":26,"value":28307}," QiskitRuntimeService, SamplerV2 ",{"type":21,"tag":322,"props":28309,"children":28310},{"style":3638},[28311],{"type":26,"value":4715},{"type":21,"tag":322,"props":28313,"children":28314},{"style":329},[28315],{"type":26,"value":28316}," Sampler\n",{"type":21,"tag":322,"props":28318,"children":28319},{"class":324,"line":407},[28320],{"type":21,"tag":322,"props":28321,"children":28322},{"emptyLinePlaceholder":3662},[28323],{"type":26,"value":3665},{"type":21,"tag":322,"props":28325,"children":28326},{"class":324,"line":431},[28327],{"type":21,"tag":322,"props":28328,"children":28329},{"style":3888},[28330],{"type":26,"value":28331},"# Build a Bell state\n",{"type":21,"tag":322,"props":28333,"children":28334},{"class":324,"line":453},[28335,28339,28343,28347,28351,28355,28359],{"type":21,"tag":322,"props":28336,"children":28337},{"style":329},[28338],{"type":26,"value":7819},{"type":21,"tag":322,"props":28340,"children":28341},{"style":3638},[28342],{"type":26,"value":3678},{"type":21,"tag":322,"props":28344,"children":28345},{"style":329},[28346],{"type":26,"value":5575},{"type":21,"tag":322,"props":28348,"children":28349},{"style":348},[28350],{"type":26,"value":423},{"type":21,"tag":322,"props":28352,"children":28353},{"style":329},[28354],{"type":26,"value":367},{"type":21,"tag":322,"props":28356,"children":28357},{"style":348},[28358],{"type":26,"value":423},{"type":21,"tag":322,"props":28360,"children":28361},{"style":329},[28362],{"type":26,"value":3738},{"type":21,"tag":322,"props":28364,"children":28365},{"class":324,"line":474},[28366,28370,28374],{"type":21,"tag":322,"props":28367,"children":28368},{"style":329},[28369],{"type":26,"value":7835},{"type":21,"tag":322,"props":28371,"children":28372},{"style":348},[28373],{"type":26,"value":5605},{"type":21,"tag":322,"props":28375,"children":28376},{"style":329},[28377],{"type":26,"value":3738},{"type":21,"tag":322,"props":28379,"children":28380},{"class":324,"line":496},[28381,28385,28389,28393,28397],{"type":21,"tag":322,"props":28382,"children":28383},{"style":329},[28384],{"type":26,"value":20825},{"type":21,"tag":322,"props":28386,"children":28387},{"style":348},[28388],{"type":26,"value":5605},{"type":21,"tag":322,"props":28390,"children":28391},{"style":329},[28392],{"type":26,"value":367},{"type":21,"tag":322,"props":28394,"children":28395},{"style":348},[28396],{"type":26,"value":3994},{"type":21,"tag":322,"props":28398,"children":28399},{"style":329},[28400],{"type":26,"value":3738},{"type":21,"tag":322,"props":28402,"children":28403},{"class":324,"line":16},[28404,28408,28412,28416,28420,28424,28428,28432,28436],{"type":21,"tag":322,"props":28405,"children":28406},{"style":329},[28407],{"type":26,"value":21633},{"type":21,"tag":322,"props":28409,"children":28410},{"style":348},[28411],{"type":26,"value":5605},{"type":21,"tag":322,"props":28413,"children":28414},{"style":329},[28415],{"type":26,"value":367},{"type":21,"tag":322,"props":28417,"children":28418},{"style":348},[28419],{"type":26,"value":3994},{"type":21,"tag":322,"props":28421,"children":28422},{"style":329},[28423],{"type":26,"value":5660},{"type":21,"tag":322,"props":28425,"children":28426},{"style":348},[28427],{"type":26,"value":5605},{"type":21,"tag":322,"props":28429,"children":28430},{"style":329},[28431],{"type":26,"value":367},{"type":21,"tag":322,"props":28433,"children":28434},{"style":348},[28435],{"type":26,"value":3994},{"type":21,"tag":322,"props":28437,"children":28438},{"style":329},[28439],{"type":26,"value":3999},{"type":21,"tag":322,"props":28441,"children":28442},{"class":324,"line":3884},[28443],{"type":21,"tag":322,"props":28444,"children":28445},{"emptyLinePlaceholder":3662},[28446],{"type":26,"value":3665},{"type":21,"tag":322,"props":28448,"children":28449},{"class":324,"line":3894},[28450],{"type":21,"tag":322,"props":28451,"children":28452},{"style":3888},[28453],{"type":26,"value":28454},"# Connect and pick the least-busy free QPU\n",{"type":21,"tag":322,"props":28456,"children":28457},{"class":324,"line":1034},[28458,28462,28466,28470,28474,28478,28482],{"type":21,"tag":322,"props":28459,"children":28460},{"style":329},[28461],{"type":26,"value":3673},{"type":21,"tag":322,"props":28463,"children":28464},{"style":3638},[28465],{"type":26,"value":3678},{"type":21,"tag":322,"props":28467,"children":28468},{"style":329},[28469],{"type":26,"value":27851},{"type":21,"tag":322,"props":28471,"children":28472},{"style":3703},[28473],{"type":26,"value":27856},{"type":21,"tag":322,"props":28475,"children":28476},{"style":3638},[28477],{"type":26,"value":3678},{"type":21,"tag":322,"props":28479,"children":28480},{"style":359},[28481],{"type":26,"value":21163},{"type":21,"tag":322,"props":28483,"children":28484},{"style":329},[28485],{"type":26,"value":3738},{"type":21,"tag":322,"props":28487,"children":28488},{"class":324,"line":3954},[28489,28493,28497],{"type":21,"tag":322,"props":28490,"children":28491},{"style":329},[28492],{"type":26,"value":3691},{"type":21,"tag":322,"props":28494,"children":28495},{"style":3638},[28496],{"type":26,"value":3678},{"type":21,"tag":322,"props":28498,"children":28499},{"style":329},[28500],{"type":26,"value":28501}," service.least_busy(\n",{"type":21,"tag":322,"props":28503,"children":28504},{"class":324,"line":4002},[28505,28509,28513,28517],{"type":21,"tag":322,"props":28506,"children":28507},{"style":3703},[28508],{"type":26,"value":27893},{"type":21,"tag":322,"props":28510,"children":28511},{"style":3638},[28512],{"type":26,"value":3678},{"type":21,"tag":322,"props":28514,"children":28515},{"style":348},[28516],{"type":26,"value":3715},{"type":21,"tag":322,"props":28518,"children":28519},{"style":329},[28520],{"type":26,"value":428},{"type":21,"tag":322,"props":28522,"children":28523},{"class":324,"line":4044},[28524,28528,28532,28536],{"type":21,"tag":322,"props":28525,"children":28526},{"style":3703},[28527],{"type":26,"value":27913},{"type":21,"tag":322,"props":28529,"children":28530},{"style":3638},[28531],{"type":26,"value":3678},{"type":21,"tag":322,"props":28533,"children":28534},{"style":348},[28535],{"type":26,"value":3733},{"type":21,"tag":322,"props":28537,"children":28538},{"style":329},[28539],{"type":26,"value":428},{"type":21,"tag":322,"props":28541,"children":28542},{"class":324,"line":5068},[28543,28547,28551],{"type":21,"tag":322,"props":28544,"children":28545},{"style":3703},[28546],{"type":26,"value":27933},{"type":21,"tag":322,"props":28548,"children":28549},{"style":3638},[28550],{"type":26,"value":3678},{"type":21,"tag":322,"props":28552,"children":28553},{"style":348},[28554],{"type":26,"value":28555},"2\n",{"type":21,"tag":322,"props":28557,"children":28558},{"class":324,"line":5087},[28559],{"type":21,"tag":322,"props":28560,"children":28561},{"style":329},[28562],{"type":26,"value":3738},{"type":21,"tag":322,"props":28564,"children":28565},{"class":324,"line":5095},[28566,28570,28574,28578,28583,28587,28591,28595,28599],{"type":21,"tag":322,"props":28567,"children":28568},{"style":348},[28569],{"type":26,"value":3753},{"type":21,"tag":322,"props":28571,"children":28572},{"style":329},[28573],{"type":26,"value":3758},{"type":21,"tag":322,"props":28575,"children":28576},{"style":3638},[28577],{"type":26,"value":3763},{"type":21,"tag":322,"props":28579,"children":28580},{"style":359},[28581],{"type":26,"value":28582},"\"Submitting to: ",{"type":21,"tag":322,"props":28584,"children":28585},{"style":348},[28586],{"type":26,"value":3773},{"type":21,"tag":322,"props":28588,"children":28589},{"style":329},[28590],{"type":26,"value":3778},{"type":21,"tag":322,"props":28592,"children":28593},{"style":348},[28594],{"type":26,"value":3783},{"type":21,"tag":322,"props":28596,"children":28597},{"style":359},[28598],{"type":26,"value":3788},{"type":21,"tag":322,"props":28600,"children":28601},{"style":329},[28602],{"type":26,"value":3738},{"type":21,"tag":322,"props":28604,"children":28605},{"class":324,"line":5126},[28606],{"type":21,"tag":322,"props":28607,"children":28608},{"emptyLinePlaceholder":3662},[28609],{"type":26,"value":3665},{"type":21,"tag":322,"props":28611,"children":28612},{"class":324,"line":5171},[28613],{"type":21,"tag":322,"props":28614,"children":28615},{"style":3888},[28616],{"type":26,"value":28617},"# Transpile for the target device\n",{"type":21,"tag":322,"props":28619,"children":28620},{"class":324,"line":5210},[28621,28625,28629,28634,28638,28642,28647,28651,28655,28659],{"type":21,"tag":322,"props":28622,"children":28623},{"style":329},[28624],{"type":26,"value":6282},{"type":21,"tag":322,"props":28626,"children":28627},{"style":3638},[28628],{"type":26,"value":3678},{"type":21,"tag":322,"props":28630,"children":28631},{"style":329},[28632],{"type":26,"value":28633}," generate_preset_pass_manager(",{"type":21,"tag":322,"props":28635,"children":28636},{"style":3703},[28637],{"type":26,"value":18908},{"type":21,"tag":322,"props":28639,"children":28640},{"style":3638},[28641],{"type":26,"value":3678},{"type":21,"tag":322,"props":28643,"children":28644},{"style":329},[28645],{"type":26,"value":28646},"backend, ",{"type":21,"tag":322,"props":28648,"children":28649},{"style":3703},[28650],{"type":26,"value":6712},{"type":21,"tag":322,"props":28652,"children":28653},{"style":3638},[28654],{"type":26,"value":3678},{"type":21,"tag":322,"props":28656,"children":28657},{"style":348},[28658],{"type":26,"value":3994},{"type":21,"tag":322,"props":28660,"children":28661},{"style":329},[28662],{"type":26,"value":3738},{"type":21,"tag":322,"props":28664,"children":28665},{"class":324,"line":5228},[28666,28671,28675],{"type":21,"tag":322,"props":28667,"children":28668},{"style":329},[28669],{"type":26,"value":28670},"isa_qc ",{"type":21,"tag":322,"props":28672,"children":28673},{"style":3638},[28674],{"type":26,"value":3678},{"type":21,"tag":322,"props":28676,"children":28677},{"style":329},[28678],{"type":26,"value":28679}," pm.run(qc)\n",{"type":21,"tag":322,"props":28681,"children":28682},{"class":324,"line":5302},[28683],{"type":21,"tag":322,"props":28684,"children":28685},{"emptyLinePlaceholder":3662},[28686],{"type":26,"value":3665},{"type":21,"tag":322,"props":28688,"children":28689},{"class":324,"line":5330},[28690],{"type":21,"tag":322,"props":28691,"children":28692},{"style":3888},[28693],{"type":26,"value":28694},"# Submit\n",{"type":21,"tag":322,"props":28696,"children":28697},{"class":324,"line":5359},[28698,28703,28707,28712,28717,28721],{"type":21,"tag":322,"props":28699,"children":28700},{"style":329},[28701],{"type":26,"value":28702},"sampler ",{"type":21,"tag":322,"props":28704,"children":28705},{"style":3638},[28706],{"type":26,"value":3678},{"type":21,"tag":322,"props":28708,"children":28709},{"style":329},[28710],{"type":26,"value":28711}," Sampler(",{"type":21,"tag":322,"props":28713,"children":28714},{"style":3703},[28715],{"type":26,"value":28716},"mode",{"type":21,"tag":322,"props":28718,"children":28719},{"style":3638},[28720],{"type":26,"value":3678},{"type":21,"tag":322,"props":28722,"children":28723},{"style":329},[28724],{"type":26,"value":28725},"backend)\n",{"type":21,"tag":322,"props":28727,"children":28728},{"class":324,"line":5368},[28729,28734,28738,28743,28747,28751,28755],{"type":21,"tag":322,"props":28730,"children":28731},{"style":329},[28732],{"type":26,"value":28733},"job ",{"type":21,"tag":322,"props":28735,"children":28736},{"style":3638},[28737],{"type":26,"value":3678},{"type":21,"tag":322,"props":28739,"children":28740},{"style":329},[28741],{"type":26,"value":28742}," sampler.run([isa_qc], ",{"type":21,"tag":322,"props":28744,"children":28745},{"style":3703},[28746],{"type":26,"value":4560},{"type":21,"tag":322,"props":28748,"children":28749},{"style":3638},[28750],{"type":26,"value":3678},{"type":21,"tag":322,"props":28752,"children":28753},{"style":348},[28754],{"type":26,"value":489},{"type":21,"tag":322,"props":28756,"children":28757},{"style":329},[28758],{"type":26,"value":3738},{"type":21,"tag":322,"props":28760,"children":28761},{"class":324,"line":5396},[28762],{"type":21,"tag":322,"props":28763,"children":28764},{"emptyLinePlaceholder":3662},[28765],{"type":26,"value":3665},{"type":21,"tag":322,"props":28767,"children":28768},{"class":324,"line":5427},[28769,28773,28777,28781,28786,28790,28795,28799,28803],{"type":21,"tag":322,"props":28770,"children":28771},{"style":348},[28772],{"type":26,"value":3753},{"type":21,"tag":322,"props":28774,"children":28775},{"style":329},[28776],{"type":26,"value":3758},{"type":21,"tag":322,"props":28778,"children":28779},{"style":3638},[28780],{"type":26,"value":3763},{"type":21,"tag":322,"props":28782,"children":28783},{"style":359},[28784],{"type":26,"value":28785},"\"Job ID: ",{"type":21,"tag":322,"props":28787,"children":28788},{"style":348},[28789],{"type":26,"value":3773},{"type":21,"tag":322,"props":28791,"children":28792},{"style":329},[28793],{"type":26,"value":28794},"job.job_id()",{"type":21,"tag":322,"props":28796,"children":28797},{"style":348},[28798],{"type":26,"value":3783},{"type":21,"tag":322,"props":28800,"children":28801},{"style":359},[28802],{"type":26,"value":3788},{"type":21,"tag":322,"props":28804,"children":28805},{"style":329},[28806],{"type":26,"value":3738},{"type":21,"tag":322,"props":28808,"children":28809},{"class":324,"line":5449},[28810,28814,28818,28822,28827,28831,28836,28840,28844],{"type":21,"tag":322,"props":28811,"children":28812},{"style":348},[28813],{"type":26,"value":3753},{"type":21,"tag":322,"props":28815,"children":28816},{"style":329},[28817],{"type":26,"value":3758},{"type":21,"tag":322,"props":28819,"children":28820},{"style":3638},[28821],{"type":26,"value":3763},{"type":21,"tag":322,"props":28823,"children":28824},{"style":359},[28825],{"type":26,"value":28826},"\"Status: ",{"type":21,"tag":322,"props":28828,"children":28829},{"style":348},[28830],{"type":26,"value":3773},{"type":21,"tag":322,"props":28832,"children":28833},{"style":329},[28834],{"type":26,"value":28835},"job.status()",{"type":21,"tag":322,"props":28837,"children":28838},{"style":348},[28839],{"type":26,"value":3783},{"type":21,"tag":322,"props":28841,"children":28842},{"style":359},[28843],{"type":26,"value":3788},{"type":21,"tag":322,"props":28845,"children":28846},{"style":329},[28847],{"type":26,"value":3738},{"type":21,"tag":322,"props":28849,"children":28850},{"class":324,"line":5495},[28851,28855,28859],{"type":21,"tag":322,"props":28852,"children":28853},{"style":329},[28854],{"type":26,"value":19764},{"type":21,"tag":322,"props":28856,"children":28857},{"style":3638},[28858],{"type":26,"value":3678},{"type":21,"tag":322,"props":28860,"children":28861},{"style":329},[28862],{"type":26,"value":28863}," job.result()\n",{"type":21,"tag":322,"props":28865,"children":28866},{"class":324,"line":5508},[28867,28871,28876,28880],{"type":21,"tag":322,"props":28868,"children":28869},{"style":348},[28870],{"type":26,"value":3753},{"type":21,"tag":322,"props":28872,"children":28873},{"style":329},[28874],{"type":26,"value":28875},"(result[",{"type":21,"tag":322,"props":28877,"children":28878},{"style":348},[28879],{"type":26,"value":5605},{"type":21,"tag":322,"props":28881,"children":28882},{"style":329},[28883],{"type":26,"value":28884},"].data.c.get_counts())\n",{"type":21,"tag":59,"props":28886,"children":28888},{"id":28887},"キュー待ち時間を理解する",[28889],{"type":26,"value":28887},{"type":21,"tag":22,"props":28891,"children":28892},{},[28893],{"type":26,"value":28894},"キュー待ち時間は、無料プランで最もよくある不満の種です。何が起きているのかを説明します:",{"type":21,"tag":3556,"props":28896,"children":28897},{},[28898,28903,28916],{"type":21,"tag":3560,"props":28899,"children":28900},{},[28901],{"type":26,"value":28902},"あなたは世界中の他のすべての Open Plan ユーザーと QPU を共有しています",{"type":21,"tag":3560,"props":28904,"children":28905},{},[28906,28908,28914],{"type":26,"value":28907},"ジョブはシステムごとにキューに入れられます。",{"type":21,"tag":79,"props":28909,"children":28911},{"className":28910},[],[28912],{"type":26,"value":28913},"least_busy()",{"type":26,"value":28915}," は最も待ち時間の短いものを見つけるのに役立ちます",{"type":21,"tag":3560,"props":28917,"children":28918},{},[28919],{"type":26,"value":28920},"一般的な待ち時間: システムの負荷に応じて5分から2時間",{"type":21,"tag":22,"props":28922,"children":28923},{},[28924],{"type":21,"tag":71,"props":28925,"children":28926},{},[28927],{"type":26,"value":28928},"キュー待ち時間を最小限に抑えるヒント:",{"type":21,"tag":3556,"props":28930,"children":28931},{},[28932,28937,28950,28955],{"type":21,"tag":3560,"props":28933,"children":28934},{},[28935],{"type":26,"value":28936},"オフピーク時間帯に実行する(ヨーロッパの朝 = アメリカの夜)",{"type":21,"tag":3560,"props":28938,"children":28939},{},[28940,28942,28948],{"type":26,"value":28941},"特定のシステムを指定するのではなく ",{"type":21,"tag":79,"props":28943,"children":28945},{"className":28944},[],[28946],{"type":26,"value":28947},"service.least_busy()",{"type":26,"value":28949}," を使う",{"type":21,"tag":3560,"props":28951,"children":28952},{},[28953],{"type":26,"value":28954},"開発時にはローカルで Aer シミュレータを使用し、最終検証のときだけ実機に投入する",{"type":21,"tag":3560,"props":28956,"children":28957},{},[28958],{"type":26,"value":28959},"SamplerV2 のリスト入力を使って複数の回路を1つのジョブにまとめる",{"type":21,"tag":59,"props":28961,"children":28963},{"id":28962},"代わりに無料のクラウドシミュレータを使う",[28964],{"type":26,"value":28962},{"type":21,"tag":22,"props":28966,"children":28967},{},[28968],{"type":26,"value":28969},"ほとんどの開発作業では、IBM のクラウドホスト型シミュレータは実機のキューに並ぶよりも高速です:",{"type":21,"tag":312,"props":28971,"children":28973},{"className":3626,"code":28972,"language":3628,"meta":7,"style":7},"# IBM cloud QASM simulator (up to 32 qubits, no queue)\nbackend = service.backend(\"ibmq_qasm_simulator\")\n\n# Or use the Statevector simulator (exact simulation)\nbackend = service.backend(\"simulator_statevector\")\n",[28974],{"type":21,"tag":79,"props":28975,"children":28976},{"__ignoreMap":7},[28977,28985,29010,29017,29025],{"type":21,"tag":322,"props":28978,"children":28979},{"class":324,"line":325},[28980],{"type":21,"tag":322,"props":28981,"children":28982},{"style":3888},[28983],{"type":26,"value":28984},"# IBM cloud QASM simulator (up to 32 qubits, no queue)\n",{"type":21,"tag":322,"props":28986,"children":28987},{"class":324,"line":335},[28988,28992,28996,29001,29006],{"type":21,"tag":322,"props":28989,"children":28990},{"style":329},[28991],{"type":26,"value":3691},{"type":21,"tag":322,"props":28993,"children":28994},{"style":3638},[28995],{"type":26,"value":3678},{"type":21,"tag":322,"props":28997,"children":28998},{"style":329},[28999],{"type":26,"value":29000}," service.backend(",{"type":21,"tag":322,"props":29002,"children":29003},{"style":359},[29004],{"type":26,"value":29005},"\"ibmq_qasm_simulator\"",{"type":21,"tag":322,"props":29007,"children":29008},{"style":329},[29009],{"type":26,"value":3738},{"type":21,"tag":322,"props":29011,"children":29012},{"class":324,"line":344},[29013],{"type":21,"tag":322,"props":29014,"children":29015},{"emptyLinePlaceholder":3662},[29016],{"type":26,"value":3665},{"type":21,"tag":322,"props":29018,"children":29019},{"class":324,"line":407},[29020],{"type":21,"tag":322,"props":29021,"children":29022},{"style":3888},[29023],{"type":26,"value":29024},"# Or use the Statevector simulator (exact simulation)\n",{"type":21,"tag":322,"props":29026,"children":29027},{"class":324,"line":431},[29028,29032,29036,29040,29045],{"type":21,"tag":322,"props":29029,"children":29030},{"style":329},[29031],{"type":26,"value":3691},{"type":21,"tag":322,"props":29033,"children":29034},{"style":3638},[29035],{"type":26,"value":3678},{"type":21,"tag":322,"props":29037,"children":29038},{"style":329},[29039],{"type":26,"value":29000},{"type":21,"tag":322,"props":29041,"children":29042},{"style":359},[29043],{"type":26,"value":29044},"\"simulator_statevector\"",{"type":21,"tag":322,"props":29046,"children":29047},{"style":329},[29048],{"type":26,"value":3738},{"type":21,"tag":22,"props":29050,"children":29051},{},[29052],{"type":26,"value":29053},"これらのシミュレータはキュー待ち時間がなく、即座に実行されます。実際のノイズ特性が必要な場合や、実機でベンチマークを取りたい場合には実 QPU を使用してください。",{"type":21,"tag":59,"props":29055,"children":29057},{"id":29056},"hlquantum-を-ibm-quantum-で使う",[29058],{"type":26,"value":29059},"HLQuantum を IBM Quantum で使う",{"type":21,"tag":22,"props":29061,"children":29062},{},[29063],{"type":26,"value":29064},"複数の SDK を使用している場合、HLQuantum は IBM ハードウェアへのよりすっきりしたインターフェースを提供します:",{"type":21,"tag":312,"props":29066,"children":29068},{"className":3626,"code":29067,"language":3628,"meta":7,"style":7},"import hlquantum as hlq\n\nqc = hlq.Circuit(2)\nqc.h(0).cx(0, 1).measure_all()\n\n# Run on IBM's free QPU\nresult = hlq.run(\n    qc,\n    backend=\"qiskit\",\n    device=\"ibm_sherbrooke\",  # or use \"auto\" for least busy\n    shots=1024,\n    error_mitigation=\"zne\"\n)\nprint(result)\n",[29069],{"type":21,"tag":79,"props":29070,"children":29071},{"__ignoreMap":7},[29072,29091,29098,29121,29152,29159,29167,29183,29191,29211,29238,29258,29275,29282],{"type":21,"tag":322,"props":29073,"children":29074},{"class":324,"line":325},[29075,29079,29083,29087],{"type":21,"tag":322,"props":29076,"children":29077},{"style":3638},[29078],{"type":26,"value":3651},{"type":21,"tag":322,"props":29080,"children":29081},{"style":329},[29082],{"type":26,"value":19732},{"type":21,"tag":322,"props":29084,"children":29085},{"style":3638},[29086],{"type":26,"value":4715},{"type":21,"tag":322,"props":29088,"children":29089},{"style":329},[29090],{"type":26,"value":19741},{"type":21,"tag":322,"props":29092,"children":29093},{"class":324,"line":335},[29094],{"type":21,"tag":322,"props":29095,"children":29096},{"emptyLinePlaceholder":3662},[29097],{"type":26,"value":3665},{"type":21,"tag":322,"props":29099,"children":29100},{"class":324,"line":344},[29101,29105,29109,29113,29117],{"type":21,"tag":322,"props":29102,"children":29103},{"style":329},[29104],{"type":26,"value":7819},{"type":21,"tag":322,"props":29106,"children":29107},{"style":3638},[29108],{"type":26,"value":3678},{"type":21,"tag":322,"props":29110,"children":29111},{"style":329},[29112],{"type":26,"value":21793},{"type":21,"tag":322,"props":29114,"children":29115},{"style":348},[29116],{"type":26,"value":423},{"type":21,"tag":322,"props":29118,"children":29119},{"style":329},[29120],{"type":26,"value":3738},{"type":21,"tag":322,"props":29122,"children":29123},{"class":324,"line":407},[29124,29128,29132,29136,29140,29144,29148],{"type":21,"tag":322,"props":29125,"children":29126},{"style":329},[29127],{"type":26,"value":7835},{"type":21,"tag":322,"props":29129,"children":29130},{"style":348},[29131],{"type":26,"value":5605},{"type":21,"tag":322,"props":29133,"children":29134},{"style":329},[29135],{"type":26,"value":21817},{"type":21,"tag":322,"props":29137,"children":29138},{"style":348},[29139],{"type":26,"value":5605},{"type":21,"tag":322,"props":29141,"children":29142},{"style":329},[29143],{"type":26,"value":367},{"type":21,"tag":322,"props":29145,"children":29146},{"style":348},[29147],{"type":26,"value":3994},{"type":21,"tag":322,"props":29149,"children":29150},{"style":329},[29151],{"type":26,"value":21834},{"type":21,"tag":322,"props":29153,"children":29154},{"class":324,"line":431},[29155],{"type":21,"tag":322,"props":29156,"children":29157},{"emptyLinePlaceholder":3662},[29158],{"type":26,"value":3665},{"type":21,"tag":322,"props":29160,"children":29161},{"class":324,"line":453},[29162],{"type":21,"tag":322,"props":29163,"children":29164},{"style":3888},[29165],{"type":26,"value":29166},"# Run on IBM's free QPU\n",{"type":21,"tag":322,"props":29168,"children":29169},{"class":324,"line":474},[29170,29174,29178],{"type":21,"tag":322,"props":29171,"children":29172},{"style":329},[29173],{"type":26,"value":19764},{"type":21,"tag":322,"props":29175,"children":29176},{"style":3638},[29177],{"type":26,"value":3678},{"type":21,"tag":322,"props":29179,"children":29180},{"style":329},[29181],{"type":26,"value":29182}," hlq.run(\n",{"type":21,"tag":322,"props":29184,"children":29185},{"class":324,"line":496},[29186],{"type":21,"tag":322,"props":29187,"children":29188},{"style":329},[29189],{"type":26,"value":29190},"    qc,\n",{"type":21,"tag":322,"props":29192,"children":29193},{"class":324,"line":16},[29194,29198,29202,29207],{"type":21,"tag":322,"props":29195,"children":29196},{"style":3703},[29197],{"type":26,"value":23300},{"type":21,"tag":322,"props":29199,"children":29200},{"style":3638},[29201],{"type":26,"value":3678},{"type":21,"tag":322,"props":29203,"children":29204},{"style":359},[29205],{"type":26,"value":29206},"\"qiskit\"",{"type":21,"tag":322,"props":29208,"children":29209},{"style":329},[29210],{"type":26,"value":428},{"type":21,"tag":322,"props":29212,"children":29213},{"class":324,"line":3884},[29214,29219,29223,29228,29233],{"type":21,"tag":322,"props":29215,"children":29216},{"style":3703},[29217],{"type":26,"value":29218},"    device",{"type":21,"tag":322,"props":29220,"children":29221},{"style":3638},[29222],{"type":26,"value":3678},{"type":21,"tag":322,"props":29224,"children":29225},{"style":359},[29226],{"type":26,"value":29227},"\"ibm_sherbrooke\"",{"type":21,"tag":322,"props":29229,"children":29230},{"style":329},[29231],{"type":26,"value":29232},",  ",{"type":21,"tag":322,"props":29234,"children":29235},{"style":3888},[29236],{"type":26,"value":29237},"# or use \"auto\" for least busy\n",{"type":21,"tag":322,"props":29239,"children":29240},{"class":324,"line":3894},[29241,29246,29250,29254],{"type":21,"tag":322,"props":29242,"children":29243},{"style":3703},[29244],{"type":26,"value":29245},"    shots",{"type":21,"tag":322,"props":29247,"children":29248},{"style":3638},[29249],{"type":26,"value":3678},{"type":21,"tag":322,"props":29251,"children":29252},{"style":348},[29253],{"type":26,"value":489},{"type":21,"tag":322,"props":29255,"children":29256},{"style":329},[29257],{"type":26,"value":428},{"type":21,"tag":322,"props":29259,"children":29260},{"class":324,"line":1034},[29261,29266,29270],{"type":21,"tag":322,"props":29262,"children":29263},{"style":3703},[29264],{"type":26,"value":29265},"    error_mitigation",{"type":21,"tag":322,"props":29267,"children":29268},{"style":3638},[29269],{"type":26,"value":3678},{"type":21,"tag":322,"props":29271,"children":29272},{"style":359},[29273],{"type":26,"value":29274},"\"zne\"\n",{"type":21,"tag":322,"props":29276,"children":29277},{"class":324,"line":3954},[29278],{"type":21,"tag":322,"props":29279,"children":29280},{"style":329},[29281],{"type":26,"value":3738},{"type":21,"tag":322,"props":29283,"children":29284},{"class":324,"line":4002},[29285,29289],{"type":21,"tag":322,"props":29286,"children":29287},{"style":348},[29288],{"type":26,"value":3753},{"type":21,"tag":322,"props":29290,"children":29291},{"style":329},[29292],{"type":26,"value":29293},"(result)\n",{"type":21,"tag":22,"props":29295,"children":29296},{},[29297,29299,29304],{"type":26,"value":29298},"HLQuantum はトランスパイル、ジョブの投入、結果の解析を自動的に処理します。高度な設定については ",{"type":21,"tag":29,"props":29300,"children":29301},{"href":7528},[29302],{"type":26,"value":29303},"Qiskit SDK ガイド",{"type":26,"value":29305}," の全文を参照してください。",{"type":21,"tag":703,"props":29307,"children":29308},{},[29309],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":29311},[29312,29313,29314,29315,29321,29322,29323],{"id":27530,"depth":335,"text":27530},{"id":27609,"depth":335,"text":27612},{"id":27684,"depth":335,"text":27687},{"id":28047,"depth":335,"text":28050,"children":29316},[29317,29318,29319,29320],{"id":28053,"depth":344,"text":28056},{"id":28069,"depth":344,"text":28072},{"id":28102,"depth":344,"text":28105},{"id":28237,"depth":344,"text":28240},{"id":28887,"depth":335,"text":28887},{"id":28962,"depth":335,"text":28962},{"id":29056,"depth":335,"text":29059},"content:ja:blog:ibm-quantum-free-tier-2026.md","ja\u002Fblog\u002Fibm-quantum-free-tier-2026.md","ja\u002Fblog\u002Fibm-quantum-free-tier-2026",{"_path":29328,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":29329,"description":29330,"date":27518,"author":11,"tags":29331,"readingTime":3894,"body":29332,"_type":715,"_id":32062,"_source":717,"_file":32063,"_stem":32064,"_extension":720},"\u002Fja\u002Fblog\u002Fqaoa-max-cut-tutorial","QAOA徹底解説：Max-Cutをステップ・バイ・ステップで解く","量子近似最適化アルゴリズム（QAOA）をMax-Cutグラフ問題に適用する実践的な解説 — Qiskit、PennyLane、HLQuantumを使った完全なPythonコード付き。",[7585,14,1962,14223],{"type":18,"children":29333,"toc":32052},[29334,29339,29344,29350,29355,29360,29368,29387,29393,29398,29435,29463,29469,30608,30628,30633,30638,31236,31241,31249,31267,31273,31278,31576,31581,31786,31791,31811,31905,31932,31938,31946,31964,31972,31985,32002,32007,32048],{"type":21,"tag":22,"props":29335,"children":29336},{},[29337],{"type":26,"value":29338},"QAOA（量子近似最適化アルゴリズム）は、最も注目を集めている近未来の量子アルゴリズムの一つであり、それには正当な理由があります。組合せ最適化問題における量子優位性の真の候補であり、今日のNISQハードウェア上で動作し、しかも午後の数時間で実装できるほど手軽です。",{"type":21,"tag":22,"props":29340,"children":29341},{},[29342],{"type":26,"value":29343},"このガイドでは、Max-Cutを例題として、QAOAの完全なパイプラインをゼロから構築します。読み終える頃には、QAOAが実際に何をするのか、どう実装するのか、そしていつ使う価値があるのかを理解できるでしょう。",{"type":21,"tag":59,"props":29345,"children":29347},{"id":29346},"max-cutとは何か",[29348],{"type":26,"value":29349},"Max-Cutとは何か",{"type":21,"tag":22,"props":29351,"children":29352},{},[29353],{"type":26,"value":29354},"Max-Cutはグラフ分割問題です。与えられたグラフに対して、ノードを2つのグループ（0と1）に分け、グループ間をまたぐエッジの数を最大化します。一般にはNP困難であり、VLSIチップ設計、クラスタリング、ネットワーク解析などに現れ、量子ハードウェアにきれいにマッピングできます。",{"type":21,"tag":22,"props":29356,"children":29357},{},[29358],{"type":26,"value":29359},"こちらはシンプルな4ノードのグラフです：",{"type":21,"tag":312,"props":29361,"children":29363},{"code":29362},"0 --- 1\n|   \\ |\n3 --- 2\n",[29364],{"type":21,"tag":79,"props":29365,"children":29366},{"__ignoreMap":7},[29367],{"type":26,"value":29362},{"type":21,"tag":22,"props":29369,"children":29370},{},[29371,29373,29379,29380,29385],{"type":26,"value":29372},"このグラフの最大カットは4です（エッジ0-1、1-2、2-3、3-0がすべてまたぐ）。私たちの目標は、これを達成するビット列 ",{"type":21,"tag":79,"props":29374,"children":29376},{"className":29375},[],[29377],{"type":26,"value":29378},"0101",{"type":26,"value":19246},{"type":21,"tag":79,"props":29381,"children":29383},{"className":29382},[],[29384],{"type":26,"value":17869},{"type":26,"value":29386}," を見つけることです。",{"type":21,"tag":59,"props":29388,"children":29390},{"id":29389},"qaoaの発想",[29391],{"type":26,"value":29392},"QAOAの発想",{"type":21,"tag":22,"props":29394,"children":29395},{},[29396],{"type":26,"value":29397},"QAOAは変分ハイブリッドアルゴリズムです。2つのパラメータ化された演算を交互に適用します：",{"type":21,"tag":7444,"props":29399,"children":29400},{},[29401,29418],{"type":21,"tag":3560,"props":29402,"children":29403},{},[29404,29409,29410,29416],{"type":21,"tag":71,"props":29405,"children":29406},{},[29407],{"type":26,"value":29408},"問題ユニタリ",{"type":26,"value":1901},{"type":21,"tag":79,"props":29411,"children":29413},{"className":29412},[],[29414],{"type":26,"value":29415},"U_C(γ)",{"type":26,"value":29417}," — コスト関数（エッジのまたぎ）を位相キックバックとしてエンコードする",{"type":21,"tag":3560,"props":29419,"children":29420},{},[29421,29426,29427,29433],{"type":21,"tag":71,"props":29422,"children":29423},{},[29424],{"type":26,"value":29425},"ミキサーユニタリ",{"type":26,"value":1901},{"type":21,"tag":79,"props":29428,"children":29430},{"className":29429},[],[29431],{"type":26,"value":29432},"U_B(β)",{"type":26,"value":29434}," — 解の間を混ぜ合わせる（X回転）",{"type":21,"tag":22,"props":29436,"children":29437},{},[29438,29440,29445,29447,29453,29455,29461],{"type":26,"value":29439},"これらの交互のユニタリを ",{"type":21,"tag":79,"props":29441,"children":29443},{"className":29442},[],[29444],{"type":26,"value":22},{"type":26,"value":29446}," 層重ねることで、QAOAは測定すると良い解が得られやすい量子状態を準備します。古典最適化器が ",{"type":21,"tag":79,"props":29448,"children":29450},{"className":29449},[],[29451],{"type":26,"value":29452},"2p",{"type":26,"value":29454}," 個の角度 ",{"type":21,"tag":79,"props":29456,"children":29458},{"className":29457},[],[29459],{"type":26,"value":29460},"(γ, β)",{"type":26,"value":29462}," を調整し、期待カット値を最大化します。",{"type":21,"tag":59,"props":29464,"children":29466},{"id":29465},"pennylaneでqaoaを構築する",[29467],{"type":26,"value":29468},"PennyLaneでQAOAを構築する",{"type":21,"tag":312,"props":29470,"children":29472},{"code":29471,"language":3628,"meta":7,"className":3626,"style":7},"import pennylane as qml\nimport numpy as np\nfrom scipy.optimize import minimize\n\n# Graph edges\nedges = [(0, 1), (1, 2), (2, 3), (3, 0)]\nn_qubits = 4\n\ndev = qml.device(\"default.qubit\", wires=n_qubits)\n\ndef cost_unitary(gamma):\n    \"\"\"Encode the Max-Cut cost into phases.\"\"\"\n    for u, v in edges:\n        qml.CNOT(wires=[u, v])\n        qml.RZ(2 * gamma, wires=v)\n        qml.CNOT(wires=[u, v])\n\ndef mixer_unitary(beta):\n    \"\"\"Apply X-rotation mixer to all qubits.\"\"\"\n    for i in range(n_qubits):\n        qml.RX(2 * beta, wires=i)\n\n@qml.qnode(dev)\ndef qaoa_circuit(params, p=1):\n    gammas = params[:p]\n    betas = params[p:]\n    # Start in uniform superposition\n    for i in range(n_qubits):\n        qml.Hadamard(wires=i)\n    # Alternate p layers\n    for layer in range(p):\n        cost_unitary(gammas[layer])\n        mixer_unitary(betas[layer])\n    return qml.probs(wires=range(n_qubits))\n\ndef expected_cut(params, p=1):\n    \"\"\"Classical cost function: expected number of cut edges.\"\"\"\n    probs = qaoa_circuit(params, p)\n    total = 0.0\n    for bitstring_idx, prob in enumerate(probs):\n        bits = format(bitstring_idx, f'0{n_qubits}b')\n        cut = sum(1 for u, v in edges if bits[u] != bits[v])\n        total += prob * cut\n    return -total  # Minimize negative cut\n\n# Optimize p=1 QAOA\np = 1\ninit_params = np.random.uniform(0, np.pi, 2 * p)\nresult = minimize(expected_cut, init_params, method='COBYLA',\n                  options={'maxiter': 200})\n\nprint(f\"Optimized params: {result.x}\")\nprint(f\"Expected cut: {-result.fun:.3f} \u002F 4.0 maximum\")\n",[29473],{"type":21,"tag":79,"props":29474,"children":29475},{"__ignoreMap":7},[29476,29495,29514,29535,29542,29550,29633,29648,29655,29690,29697,29714,29722,29743,29763,29797,29816,29823,29840,29848,29871,29904,29911,29922,29951,29968,29985,29993,30016,30036,30044,30069,30077,30085,30112,30119,30147,30155,30172,30189,30214,30264,30326,30353,30374,30381,30389,30405,30443,30476,30510,30517,30558],{"type":21,"tag":322,"props":29477,"children":29478},{"class":324,"line":325},[29479,29483,29487,29491],{"type":21,"tag":322,"props":29480,"children":29481},{"style":3638},[29482],{"type":26,"value":3651},{"type":21,"tag":322,"props":29484,"children":29485},{"style":329},[29486],{"type":26,"value":14530},{"type":21,"tag":322,"props":29488,"children":29489},{"style":3638},[29490],{"type":26,"value":4715},{"type":21,"tag":322,"props":29492,"children":29493},{"style":329},[29494],{"type":26,"value":14539},{"type":21,"tag":322,"props":29496,"children":29497},{"class":324,"line":335},[29498,29502,29506,29510],{"type":21,"tag":322,"props":29499,"children":29500},{"style":3638},[29501],{"type":26,"value":3651},{"type":21,"tag":322,"props":29503,"children":29504},{"style":329},[29505],{"type":26,"value":4710},{"type":21,"tag":322,"props":29507,"children":29508},{"style":3638},[29509],{"type":26,"value":4715},{"type":21,"tag":322,"props":29511,"children":29512},{"style":329},[29513],{"type":26,"value":4720},{"type":21,"tag":322,"props":29515,"children":29516},{"class":324,"line":344},[29517,29521,29526,29530],{"type":21,"tag":322,"props":29518,"children":29519},{"style":3638},[29520],{"type":26,"value":3641},{"type":21,"tag":322,"props":29522,"children":29523},{"style":329},[29524],{"type":26,"value":29525}," scipy.optimize ",{"type":21,"tag":322,"props":29527,"children":29528},{"style":3638},[29529],{"type":26,"value":3651},{"type":21,"tag":322,"props":29531,"children":29532},{"style":329},[29533],{"type":26,"value":29534}," minimize\n",{"type":21,"tag":322,"props":29536,"children":29537},{"class":324,"line":407},[29538],{"type":21,"tag":322,"props":29539,"children":29540},{"emptyLinePlaceholder":3662},[29541],{"type":26,"value":3665},{"type":21,"tag":322,"props":29543,"children":29544},{"class":324,"line":431},[29545],{"type":21,"tag":322,"props":29546,"children":29547},{"style":3888},[29548],{"type":26,"value":29549},"# Graph edges\n",{"type":21,"tag":322,"props":29551,"children":29552},{"class":324,"line":453},[29553,29558,29562,29567,29571,29575,29579,29584,29588,29592,29596,29600,29604,29608,29612,29616,29620,29624,29628],{"type":21,"tag":322,"props":29554,"children":29555},{"style":329},[29556],{"type":26,"value":29557},"edges ",{"type":21,"tag":322,"props":29559,"children":29560},{"style":3638},[29561],{"type":26,"value":3678},{"type":21,"tag":322,"props":29563,"children":29564},{"style":329},[29565],{"type":26,"value":29566}," [(",{"type":21,"tag":322,"props":29568,"children":29569},{"style":348},[29570],{"type":26,"value":5605},{"type":21,"tag":322,"props":29572,"children":29573},{"style":329},[29574],{"type":26,"value":367},{"type":21,"tag":322,"props":29576,"children":29577},{"style":348},[29578],{"type":26,"value":3994},{"type":21,"tag":322,"props":29580,"children":29581},{"style":329},[29582],{"type":26,"value":29583},"), (",{"type":21,"tag":322,"props":29585,"children":29586},{"style":348},[29587],{"type":26,"value":3994},{"type":21,"tag":322,"props":29589,"children":29590},{"style":329},[29591],{"type":26,"value":367},{"type":21,"tag":322,"props":29593,"children":29594},{"style":348},[29595],{"type":26,"value":423},{"type":21,"tag":322,"props":29597,"children":29598},{"style":329},[29599],{"type":26,"value":29583},{"type":21,"tag":322,"props":29601,"children":29602},{"style":348},[29603],{"type":26,"value":423},{"type":21,"tag":322,"props":29605,"children":29606},{"style":329},[29607],{"type":26,"value":367},{"type":21,"tag":322,"props":29609,"children":29610},{"style":348},[29611],{"type":26,"value":6948},{"type":21,"tag":322,"props":29613,"children":29614},{"style":329},[29615],{"type":26,"value":29583},{"type":21,"tag":322,"props":29617,"children":29618},{"style":348},[29619],{"type":26,"value":6948},{"type":21,"tag":322,"props":29621,"children":29622},{"style":329},[29623],{"type":26,"value":367},{"type":21,"tag":322,"props":29625,"children":29626},{"style":348},[29627],{"type":26,"value":5605},{"type":21,"tag":322,"props":29629,"children":29630},{"style":329},[29631],{"type":26,"value":29632},")]\n",{"type":21,"tag":322,"props":29634,"children":29635},{"class":324,"line":474},[29636,29640,29644],{"type":21,"tag":322,"props":29637,"children":29638},{"style":329},[29639],{"type":26,"value":14623},{"type":21,"tag":322,"props":29641,"children":29642},{"style":3638},[29643],{"type":26,"value":3678},{"type":21,"tag":322,"props":29645,"children":29646},{"style":348},[29647],{"type":26,"value":12905},{"type":21,"tag":322,"props":29649,"children":29650},{"class":324,"line":496},[29651],{"type":21,"tag":322,"props":29652,"children":29653},{"emptyLinePlaceholder":3662},[29654],{"type":26,"value":3665},{"type":21,"tag":322,"props":29656,"children":29657},{"class":324,"line":16},[29658,29662,29666,29670,29674,29678,29682,29686],{"type":21,"tag":322,"props":29659,"children":29660},{"style":329},[29661],{"type":26,"value":14663},{"type":21,"tag":322,"props":29663,"children":29664},{"style":3638},[29665],{"type":26,"value":3678},{"type":21,"tag":322,"props":29667,"children":29668},{"style":329},[29669],{"type":26,"value":14672},{"type":21,"tag":322,"props":29671,"children":29672},{"style":359},[29673],{"type":26,"value":14677},{"type":21,"tag":322,"props":29675,"children":29676},{"style":329},[29677],{"type":26,"value":367},{"type":21,"tag":322,"props":29679,"children":29680},{"style":3703},[29681],{"type":26,"value":14686},{"type":21,"tag":322,"props":29683,"children":29684},{"style":3638},[29685],{"type":26,"value":3678},{"type":21,"tag":322,"props":29687,"children":29688},{"style":329},[29689],{"type":26,"value":14695},{"type":21,"tag":322,"props":29691,"children":29692},{"class":324,"line":3884},[29693],{"type":21,"tag":322,"props":29694,"children":29695},{"emptyLinePlaceholder":3662},[29696],{"type":26,"value":3665},{"type":21,"tag":322,"props":29698,"children":29699},{"class":324,"line":3894},[29700,29704,29709],{"type":21,"tag":322,"props":29701,"children":29702},{"style":3638},[29703],{"type":26,"value":4798},{"type":21,"tag":322,"props":29705,"children":29706},{"style":4801},[29707],{"type":26,"value":29708}," cost_unitary",{"type":21,"tag":322,"props":29710,"children":29711},{"style":329},[29712],{"type":26,"value":29713},"(gamma):\n",{"type":21,"tag":322,"props":29715,"children":29716},{"class":324,"line":1034},[29717],{"type":21,"tag":322,"props":29718,"children":29719},{"style":359},[29720],{"type":26,"value":29721},"    \"\"\"Encode the Max-Cut cost into phases.\"\"\"\n",{"type":21,"tag":322,"props":29723,"children":29724},{"class":324,"line":3954},[29725,29729,29734,29738],{"type":21,"tag":322,"props":29726,"children":29727},{"style":3638},[29728],{"type":26,"value":5177},{"type":21,"tag":322,"props":29730,"children":29731},{"style":329},[29732],{"type":26,"value":29733}," u, v ",{"type":21,"tag":322,"props":29735,"children":29736},{"style":3638},[29737],{"type":26,"value":3941},{"type":21,"tag":322,"props":29739,"children":29740},{"style":329},[29741],{"type":26,"value":29742}," edges:\n",{"type":21,"tag":322,"props":29744,"children":29745},{"class":324,"line":4002},[29746,29750,29754,29758],{"type":21,"tag":322,"props":29747,"children":29748},{"style":329},[29749],{"type":26,"value":27081},{"type":21,"tag":322,"props":29751,"children":29752},{"style":3703},[29753],{"type":26,"value":14686},{"type":21,"tag":322,"props":29755,"children":29756},{"style":3638},[29757],{"type":26,"value":3678},{"type":21,"tag":322,"props":29759,"children":29760},{"style":329},[29761],{"type":26,"value":29762},"[u, v])\n",{"type":21,"tag":322,"props":29764,"children":29765},{"class":324,"line":4044},[29766,29771,29775,29779,29784,29788,29792],{"type":21,"tag":322,"props":29767,"children":29768},{"style":329},[29769],{"type":26,"value":29770},"        qml.RZ(",{"type":21,"tag":322,"props":29772,"children":29773},{"style":348},[29774],{"type":26,"value":423},{"type":21,"tag":322,"props":29776,"children":29777},{"style":3638},[29778],{"type":26,"value":8782},{"type":21,"tag":322,"props":29780,"children":29781},{"style":329},[29782],{"type":26,"value":29783}," gamma, ",{"type":21,"tag":322,"props":29785,"children":29786},{"style":3703},[29787],{"type":26,"value":14686},{"type":21,"tag":322,"props":29789,"children":29790},{"style":3638},[29791],{"type":26,"value":3678},{"type":21,"tag":322,"props":29793,"children":29794},{"style":329},[29795],{"type":26,"value":29796},"v)\n",{"type":21,"tag":322,"props":29798,"children":29799},{"class":324,"line":5068},[29800,29804,29808,29812],{"type":21,"tag":322,"props":29801,"children":29802},{"style":329},[29803],{"type":26,"value":27081},{"type":21,"tag":322,"props":29805,"children":29806},{"style":3703},[29807],{"type":26,"value":14686},{"type":21,"tag":322,"props":29809,"children":29810},{"style":3638},[29811],{"type":26,"value":3678},{"type":21,"tag":322,"props":29813,"children":29814},{"style":329},[29815],{"type":26,"value":29762},{"type":21,"tag":322,"props":29817,"children":29818},{"class":324,"line":5087},[29819],{"type":21,"tag":322,"props":29820,"children":29821},{"emptyLinePlaceholder":3662},[29822],{"type":26,"value":3665},{"type":21,"tag":322,"props":29824,"children":29825},{"class":324,"line":5095},[29826,29830,29835],{"type":21,"tag":322,"props":29827,"children":29828},{"style":3638},[29829],{"type":26,"value":4798},{"type":21,"tag":322,"props":29831,"children":29832},{"style":4801},[29833],{"type":26,"value":29834}," mixer_unitary",{"type":21,"tag":322,"props":29836,"children":29837},{"style":329},[29838],{"type":26,"value":29839},"(beta):\n",{"type":21,"tag":322,"props":29841,"children":29842},{"class":324,"line":5126},[29843],{"type":21,"tag":322,"props":29844,"children":29845},{"style":359},[29846],{"type":26,"value":29847},"    \"\"\"Apply X-rotation mixer to all qubits.\"\"\"\n",{"type":21,"tag":322,"props":29849,"children":29850},{"class":324,"line":5171},[29851,29855,29859,29863,29867],{"type":21,"tag":322,"props":29852,"children":29853},{"style":3638},[29854],{"type":26,"value":5177},{"type":21,"tag":322,"props":29856,"children":29857},{"style":329},[29858],{"type":26,"value":5182},{"type":21,"tag":322,"props":29860,"children":29861},{"style":3638},[29862],{"type":26,"value":3941},{"type":21,"tag":322,"props":29864,"children":29865},{"style":348},[29866],{"type":26,"value":3946},{"type":21,"tag":322,"props":29868,"children":29869},{"style":329},[29870],{"type":26,"value":27016},{"type":21,"tag":322,"props":29872,"children":29873},{"class":324,"line":5210},[29874,29879,29883,29887,29892,29896,29900],{"type":21,"tag":322,"props":29875,"children":29876},{"style":329},[29877],{"type":26,"value":29878},"        qml.RX(",{"type":21,"tag":322,"props":29880,"children":29881},{"style":348},[29882],{"type":26,"value":423},{"type":21,"tag":322,"props":29884,"children":29885},{"style":3638},[29886],{"type":26,"value":8782},{"type":21,"tag":322,"props":29888,"children":29889},{"style":329},[29890],{"type":26,"value":29891}," beta, ",{"type":21,"tag":322,"props":29893,"children":29894},{"style":3703},[29895],{"type":26,"value":14686},{"type":21,"tag":322,"props":29897,"children":29898},{"style":3638},[29899],{"type":26,"value":3678},{"type":21,"tag":322,"props":29901,"children":29902},{"style":329},[29903],{"type":26,"value":27037},{"type":21,"tag":322,"props":29905,"children":29906},{"class":324,"line":5228},[29907],{"type":21,"tag":322,"props":29908,"children":29909},{"emptyLinePlaceholder":3662},[29910],{"type":26,"value":3665},{"type":21,"tag":322,"props":29912,"children":29913},{"class":324,"line":5302},[29914,29918],{"type":21,"tag":322,"props":29915,"children":29916},{"style":4801},[29917],{"type":26,"value":14710},{"type":21,"tag":322,"props":29919,"children":29920},{"style":329},[29921],{"type":26,"value":14715},{"type":21,"tag":322,"props":29923,"children":29924},{"class":324,"line":5330},[29925,29929,29934,29939,29943,29947],{"type":21,"tag":322,"props":29926,"children":29927},{"style":3638},[29928],{"type":26,"value":4798},{"type":21,"tag":322,"props":29930,"children":29931},{"style":4801},[29932],{"type":26,"value":29933}," qaoa_circuit",{"type":21,"tag":322,"props":29935,"children":29936},{"style":329},[29937],{"type":26,"value":29938},"(params, p",{"type":21,"tag":322,"props":29940,"children":29941},{"style":3638},[29942],{"type":26,"value":3678},{"type":21,"tag":322,"props":29944,"children":29945},{"style":348},[29946],{"type":26,"value":3994},{"type":21,"tag":322,"props":29948,"children":29949},{"style":329},[29950],{"type":26,"value":4851},{"type":21,"tag":322,"props":29952,"children":29953},{"class":324,"line":5359},[29954,29959,29963],{"type":21,"tag":322,"props":29955,"children":29956},{"style":329},[29957],{"type":26,"value":29958},"    gammas ",{"type":21,"tag":322,"props":29960,"children":29961},{"style":3638},[29962],{"type":26,"value":3678},{"type":21,"tag":322,"props":29964,"children":29965},{"style":329},[29966],{"type":26,"value":29967}," params[:p]\n",{"type":21,"tag":322,"props":29969,"children":29970},{"class":324,"line":5368},[29971,29976,29980],{"type":21,"tag":322,"props":29972,"children":29973},{"style":329},[29974],{"type":26,"value":29975},"    betas ",{"type":21,"tag":322,"props":29977,"children":29978},{"style":3638},[29979],{"type":26,"value":3678},{"type":21,"tag":322,"props":29981,"children":29982},{"style":329},[29983],{"type":26,"value":29984}," params[p:]\n",{"type":21,"tag":322,"props":29986,"children":29987},{"class":324,"line":5396},[29988],{"type":21,"tag":322,"props":29989,"children":29990},{"style":3888},[29991],{"type":26,"value":29992},"    # Start in uniform superposition\n",{"type":21,"tag":322,"props":29994,"children":29995},{"class":324,"line":5427},[29996,30000,30004,30008,30012],{"type":21,"tag":322,"props":29997,"children":29998},{"style":3638},[29999],{"type":26,"value":5177},{"type":21,"tag":322,"props":30001,"children":30002},{"style":329},[30003],{"type":26,"value":5182},{"type":21,"tag":322,"props":30005,"children":30006},{"style":3638},[30007],{"type":26,"value":3941},{"type":21,"tag":322,"props":30009,"children":30010},{"style":348},[30011],{"type":26,"value":3946},{"type":21,"tag":322,"props":30013,"children":30014},{"style":329},[30015],{"type":26,"value":27016},{"type":21,"tag":322,"props":30017,"children":30018},{"class":324,"line":5449},[30019,30024,30028,30032],{"type":21,"tag":322,"props":30020,"children":30021},{"style":329},[30022],{"type":26,"value":30023},"        qml.Hadamard(",{"type":21,"tag":322,"props":30025,"children":30026},{"style":3703},[30027],{"type":26,"value":14686},{"type":21,"tag":322,"props":30029,"children":30030},{"style":3638},[30031],{"type":26,"value":3678},{"type":21,"tag":322,"props":30033,"children":30034},{"style":329},[30035],{"type":26,"value":27037},{"type":21,"tag":322,"props":30037,"children":30038},{"class":324,"line":5495},[30039],{"type":21,"tag":322,"props":30040,"children":30041},{"style":3888},[30042],{"type":26,"value":30043},"    # Alternate p layers\n",{"type":21,"tag":322,"props":30045,"children":30046},{"class":324,"line":5508},[30047,30051,30056,30060,30064],{"type":21,"tag":322,"props":30048,"children":30049},{"style":3638},[30050],{"type":26,"value":5177},{"type":21,"tag":322,"props":30052,"children":30053},{"style":329},[30054],{"type":26,"value":30055}," layer ",{"type":21,"tag":322,"props":30057,"children":30058},{"style":3638},[30059],{"type":26,"value":3941},{"type":21,"tag":322,"props":30061,"children":30062},{"style":348},[30063],{"type":26,"value":3946},{"type":21,"tag":322,"props":30065,"children":30066},{"style":329},[30067],{"type":26,"value":30068},"(p):\n",{"type":21,"tag":322,"props":30070,"children":30071},{"class":324,"line":5516},[30072],{"type":21,"tag":322,"props":30073,"children":30074},{"style":329},[30075],{"type":26,"value":30076},"        cost_unitary(gammas[layer])\n",{"type":21,"tag":322,"props":30078,"children":30079},{"class":324,"line":5534},[30080],{"type":21,"tag":322,"props":30081,"children":30082},{"style":329},[30083],{"type":26,"value":30084},"        mixer_unitary(betas[layer])\n",{"type":21,"tag":322,"props":30086,"children":30087},{"class":324,"line":5552},[30088,30092,30096,30100,30104,30108],{"type":21,"tag":322,"props":30089,"children":30090},{"style":3638},[30091],{"type":26,"value":5023},{"type":21,"tag":322,"props":30093,"children":30094},{"style":329},[30095],{"type":26,"value":22251},{"type":21,"tag":322,"props":30097,"children":30098},{"style":3703},[30099],{"type":26,"value":14686},{"type":21,"tag":322,"props":30101,"children":30102},{"style":3638},[30103],{"type":26,"value":3678},{"type":21,"tag":322,"props":30105,"children":30106},{"style":348},[30107],{"type":26,"value":5379},{"type":21,"tag":322,"props":30109,"children":30110},{"style":329},[30111],{"type":26,"value":14823},{"type":21,"tag":322,"props":30113,"children":30114},{"class":324,"line":5560},[30115],{"type":21,"tag":322,"props":30116,"children":30117},{"emptyLinePlaceholder":3662},[30118],{"type":26,"value":3665},{"type":21,"tag":322,"props":30120,"children":30121},{"class":324,"line":5594},[30122,30126,30131,30135,30139,30143],{"type":21,"tag":322,"props":30123,"children":30124},{"style":3638},[30125],{"type":26,"value":4798},{"type":21,"tag":322,"props":30127,"children":30128},{"style":4801},[30129],{"type":26,"value":30130}," expected_cut",{"type":21,"tag":322,"props":30132,"children":30133},{"style":329},[30134],{"type":26,"value":29938},{"type":21,"tag":322,"props":30136,"children":30137},{"style":3638},[30138],{"type":26,"value":3678},{"type":21,"tag":322,"props":30140,"children":30141},{"style":348},[30142],{"type":26,"value":3994},{"type":21,"tag":322,"props":30144,"children":30145},{"style":329},[30146],{"type":26,"value":4851},{"type":21,"tag":322,"props":30148,"children":30149},{"class":324,"line":5612},[30150],{"type":21,"tag":322,"props":30151,"children":30152},{"style":359},[30153],{"type":26,"value":30154},"    \"\"\"Classical cost function: expected number of cut edges.\"\"\"\n",{"type":21,"tag":322,"props":30156,"children":30157},{"class":324,"line":5637},[30158,30163,30167],{"type":21,"tag":322,"props":30159,"children":30160},{"style":329},[30161],{"type":26,"value":30162},"    probs ",{"type":21,"tag":322,"props":30164,"children":30165},{"style":3638},[30166],{"type":26,"value":3678},{"type":21,"tag":322,"props":30168,"children":30169},{"style":329},[30170],{"type":26,"value":30171}," qaoa_circuit(params, p)\n",{"type":21,"tag":322,"props":30173,"children":30174},{"class":324,"line":5679},[30175,30180,30184],{"type":21,"tag":322,"props":30176,"children":30177},{"style":329},[30178],{"type":26,"value":30179},"    total ",{"type":21,"tag":322,"props":30181,"children":30182},{"style":3638},[30183],{"type":26,"value":3678},{"type":21,"tag":322,"props":30185,"children":30186},{"style":348},[30187],{"type":26,"value":30188}," 0.0\n",{"type":21,"tag":322,"props":30190,"children":30191},{"class":324,"line":5687},[30192,30196,30201,30205,30209],{"type":21,"tag":322,"props":30193,"children":30194},{"style":3638},[30195],{"type":26,"value":5177},{"type":21,"tag":322,"props":30197,"children":30198},{"style":329},[30199],{"type":26,"value":30200}," bitstring_idx, prob ",{"type":21,"tag":322,"props":30202,"children":30203},{"style":3638},[30204],{"type":26,"value":3941},{"type":21,"tag":322,"props":30206,"children":30207},{"style":348},[30208],{"type":26,"value":5322},{"type":21,"tag":322,"props":30210,"children":30211},{"style":329},[30212],{"type":26,"value":30213},"(probs):\n",{"type":21,"tag":322,"props":30215,"children":30216},{"class":324,"line":5722},[30217,30221,30225,30229,30234,30238,30243,30247,30251,30255,30260],{"type":21,"tag":322,"props":30218,"children":30219},{"style":329},[30220],{"type":26,"value":5234},{"type":21,"tag":322,"props":30222,"children":30223},{"style":3638},[30224],{"type":26,"value":3678},{"type":21,"tag":322,"props":30226,"children":30227},{"style":348},[30228],{"type":26,"value":5243},{"type":21,"tag":322,"props":30230,"children":30231},{"style":329},[30232],{"type":26,"value":30233},"(bitstring_idx, ",{"type":21,"tag":322,"props":30235,"children":30236},{"style":3638},[30237],{"type":26,"value":3763},{"type":21,"tag":322,"props":30239,"children":30240},{"style":359},[30241],{"type":26,"value":30242},"'0",{"type":21,"tag":322,"props":30244,"children":30245},{"style":348},[30246],{"type":26,"value":3773},{"type":21,"tag":322,"props":30248,"children":30249},{"style":329},[30250],{"type":26,"value":15935},{"type":21,"tag":322,"props":30252,"children":30253},{"style":348},[30254],{"type":26,"value":3783},{"type":21,"tag":322,"props":30256,"children":30257},{"style":359},[30258],{"type":26,"value":30259},"b'",{"type":21,"tag":322,"props":30261,"children":30262},{"style":329},[30263],{"type":26,"value":3738},{"type":21,"tag":322,"props":30265,"children":30266},{"class":324,"line":5838},[30267,30272,30276,30280,30284,30288,30293,30297,30301,30306,30311,30316,30321],{"type":21,"tag":322,"props":30268,"children":30269},{"style":329},[30270],{"type":26,"value":30271},"        cut ",{"type":21,"tag":322,"props":30273,"children":30274},{"style":3638},[30275],{"type":26,"value":3678},{"type":21,"tag":322,"props":30277,"children":30278},{"style":348},[30279],{"type":26,"value":6757},{"type":21,"tag":322,"props":30281,"children":30282},{"style":329},[30283],{"type":26,"value":3758},{"type":21,"tag":322,"props":30285,"children":30286},{"style":348},[30287],{"type":26,"value":3994},{"type":21,"tag":322,"props":30289,"children":30290},{"style":3638},[30291],{"type":26,"value":30292}," for",{"type":21,"tag":322,"props":30294,"children":30295},{"style":329},[30296],{"type":26,"value":29733},{"type":21,"tag":322,"props":30298,"children":30299},{"style":3638},[30300],{"type":26,"value":3941},{"type":21,"tag":322,"props":30302,"children":30303},{"style":329},[30304],{"type":26,"value":30305}," edges ",{"type":21,"tag":322,"props":30307,"children":30308},{"style":3638},[30309],{"type":26,"value":30310},"if",{"type":21,"tag":322,"props":30312,"children":30313},{"style":329},[30314],{"type":26,"value":30315}," bits[u] ",{"type":21,"tag":322,"props":30317,"children":30318},{"style":3638},[30319],{"type":26,"value":30320},"!=",{"type":21,"tag":322,"props":30322,"children":30323},{"style":329},[30324],{"type":26,"value":30325}," bits[v])\n",{"type":21,"tag":322,"props":30327,"children":30328},{"class":324,"line":5846},[30329,30334,30339,30344,30348],{"type":21,"tag":322,"props":30330,"children":30331},{"style":329},[30332],{"type":26,"value":30333},"        total ",{"type":21,"tag":322,"props":30335,"children":30336},{"style":3638},[30337],{"type":26,"value":30338},"+=",{"type":21,"tag":322,"props":30340,"children":30341},{"style":329},[30342],{"type":26,"value":30343}," prob ",{"type":21,"tag":322,"props":30345,"children":30346},{"style":3638},[30347],{"type":26,"value":6830},{"type":21,"tag":322,"props":30349,"children":30350},{"style":329},[30351],{"type":26,"value":30352}," cut\n",{"type":21,"tag":322,"props":30354,"children":30355},{"class":324,"line":5891},[30356,30360,30364,30369],{"type":21,"tag":322,"props":30357,"children":30358},{"style":3638},[30359],{"type":26,"value":5023},{"type":21,"tag":322,"props":30361,"children":30362},{"style":3638},[30363],{"type":26,"value":4989},{"type":21,"tag":322,"props":30365,"children":30366},{"style":329},[30367],{"type":26,"value":30368},"total  ",{"type":21,"tag":322,"props":30370,"children":30371},{"style":3888},[30372],{"type":26,"value":30373},"# Minimize negative cut\n",{"type":21,"tag":322,"props":30375,"children":30376},{"class":324,"line":5909},[30377],{"type":21,"tag":322,"props":30378,"children":30379},{"emptyLinePlaceholder":3662},[30380],{"type":26,"value":3665},{"type":21,"tag":322,"props":30382,"children":30383},{"class":324,"line":5917},[30384],{"type":21,"tag":322,"props":30385,"children":30386},{"style":3888},[30387],{"type":26,"value":30388},"# Optimize p=1 QAOA\n",{"type":21,"tag":322,"props":30390,"children":30391},{"class":324,"line":6005},[30392,30397,30401],{"type":21,"tag":322,"props":30393,"children":30394},{"style":329},[30395],{"type":26,"value":30396},"p ",{"type":21,"tag":322,"props":30398,"children":30399},{"style":3638},[30400],{"type":26,"value":3678},{"type":21,"tag":322,"props":30402,"children":30403},{"style":348},[30404],{"type":26,"value":13231},{"type":21,"tag":322,"props":30406,"children":30407},{"class":324,"line":15510},[30408,30413,30417,30421,30425,30430,30434,30438],{"type":21,"tag":322,"props":30409,"children":30410},{"style":329},[30411],{"type":26,"value":30412},"init_params ",{"type":21,"tag":322,"props":30414,"children":30415},{"style":3638},[30416],{"type":26,"value":3678},{"type":21,"tag":322,"props":30418,"children":30419},{"style":329},[30420],{"type":26,"value":25681},{"type":21,"tag":322,"props":30422,"children":30423},{"style":348},[30424],{"type":26,"value":5605},{"type":21,"tag":322,"props":30426,"children":30427},{"style":329},[30428],{"type":26,"value":30429},", np.pi, ",{"type":21,"tag":322,"props":30431,"children":30432},{"style":348},[30433],{"type":26,"value":423},{"type":21,"tag":322,"props":30435,"children":30436},{"style":3638},[30437],{"type":26,"value":8782},{"type":21,"tag":322,"props":30439,"children":30440},{"style":329},[30441],{"type":26,"value":30442}," p)\n",{"type":21,"tag":322,"props":30444,"children":30445},{"class":324,"line":15528},[30446,30450,30454,30459,30464,30468,30472],{"type":21,"tag":322,"props":30447,"children":30448},{"style":329},[30449],{"type":26,"value":19764},{"type":21,"tag":322,"props":30451,"children":30452},{"style":3638},[30453],{"type":26,"value":3678},{"type":21,"tag":322,"props":30455,"children":30456},{"style":329},[30457],{"type":26,"value":30458}," minimize(expected_cut, init_params, ",{"type":21,"tag":322,"props":30460,"children":30461},{"style":3703},[30462],{"type":26,"value":30463},"method",{"type":21,"tag":322,"props":30465,"children":30466},{"style":3638},[30467],{"type":26,"value":3678},{"type":21,"tag":322,"props":30469,"children":30470},{"style":359},[30471],{"type":26,"value":23329},{"type":21,"tag":322,"props":30473,"children":30474},{"style":329},[30475],{"type":26,"value":428},{"type":21,"tag":322,"props":30477,"children":30478},{"class":324,"line":15546},[30479,30484,30488,30492,30497,30501,30505],{"type":21,"tag":322,"props":30480,"children":30481},{"style":3703},[30482],{"type":26,"value":30483},"                  options",{"type":21,"tag":322,"props":30485,"children":30486},{"style":3638},[30487],{"type":26,"value":3678},{"type":21,"tag":322,"props":30489,"children":30490},{"style":329},[30491],{"type":26,"value":3773},{"type":21,"tag":322,"props":30493,"children":30494},{"style":359},[30495],{"type":26,"value":30496},"'maxiter'",{"type":21,"tag":322,"props":30498,"children":30499},{"style":329},[30500],{"type":26,"value":418},{"type":21,"tag":322,"props":30502,"children":30503},{"style":348},[30504],{"type":26,"value":15068},{"type":21,"tag":322,"props":30506,"children":30507},{"style":329},[30508],{"type":26,"value":30509},"})\n",{"type":21,"tag":322,"props":30511,"children":30512},{"class":324,"line":15554},[30513],{"type":21,"tag":322,"props":30514,"children":30515},{"emptyLinePlaceholder":3662},[30516],{"type":26,"value":3665},{"type":21,"tag":322,"props":30518,"children":30519},{"class":324,"line":15588},[30520,30524,30528,30532,30537,30541,30546,30550,30554],{"type":21,"tag":322,"props":30521,"children":30522},{"style":348},[30523],{"type":26,"value":3753},{"type":21,"tag":322,"props":30525,"children":30526},{"style":329},[30527],{"type":26,"value":3758},{"type":21,"tag":322,"props":30529,"children":30530},{"style":3638},[30531],{"type":26,"value":3763},{"type":21,"tag":322,"props":30533,"children":30534},{"style":359},[30535],{"type":26,"value":30536},"\"Optimized params: ",{"type":21,"tag":322,"props":30538,"children":30539},{"style":348},[30540],{"type":26,"value":3773},{"type":21,"tag":322,"props":30542,"children":30543},{"style":329},[30544],{"type":26,"value":30545},"result.x",{"type":21,"tag":322,"props":30547,"children":30548},{"style":348},[30549],{"type":26,"value":3783},{"type":21,"tag":322,"props":30551,"children":30552},{"style":359},[30553],{"type":26,"value":3788},{"type":21,"tag":322,"props":30555,"children":30556},{"style":329},[30557],{"type":26,"value":3738},{"type":21,"tag":322,"props":30559,"children":30560},{"class":324,"line":15623},[30561,30565,30569,30573,30578,30582,30586,30591,30595,30599,30604],{"type":21,"tag":322,"props":30562,"children":30563},{"style":348},[30564],{"type":26,"value":3753},{"type":21,"tag":322,"props":30566,"children":30567},{"style":329},[30568],{"type":26,"value":3758},{"type":21,"tag":322,"props":30570,"children":30571},{"style":3638},[30572],{"type":26,"value":3763},{"type":21,"tag":322,"props":30574,"children":30575},{"style":359},[30576],{"type":26,"value":30577},"\"Expected cut: ",{"type":21,"tag":322,"props":30579,"children":30580},{"style":348},[30581],{"type":26,"value":3773},{"type":21,"tag":322,"props":30583,"children":30584},{"style":3638},[30585],{"type":26,"value":5285},{"type":21,"tag":322,"props":30587,"children":30588},{"style":329},[30589],{"type":26,"value":30590},"result.fun",{"type":21,"tag":322,"props":30592,"children":30593},{"style":3638},[30594],{"type":26,"value":15726},{"type":21,"tag":322,"props":30596,"children":30597},{"style":348},[30598],{"type":26,"value":3783},{"type":21,"tag":322,"props":30600,"children":30601},{"style":359},[30602],{"type":26,"value":30603}," \u002F 4.0 maximum\"",{"type":21,"tag":322,"props":30605,"children":30606},{"style":329},[30607],{"type":26,"value":3738},{"type":21,"tag":22,"props":30609,"children":30610},{},[30611,30613,30618,30620,30626],{"type":26,"value":30612},"これを実行すると、",{"type":21,"tag":79,"props":30614,"children":30616},{"className":30615},[],[30617],{"type":26,"value":19975},{"type":26,"value":30619}," で典型的には ",{"type":21,"tag":79,"props":30621,"children":30623},{"className":30622},[],[30624],{"type":26,"value":30625},"expected_cut ≈ 3.5",{"type":26,"value":30627}," に達します — 最大カットの87.5%です。",{"type":21,"tag":59,"props":30629,"children":30631},{"id":30630},"最良解のサンプリング",[30632],{"type":26,"value":30630},{"type":21,"tag":22,"props":30634,"children":30635},{},[30636],{"type":26,"value":30637},"最適化の後、回路をサンプリングして最も可能性の高い良い解を見つけます：",{"type":21,"tag":312,"props":30639,"children":30641},{"code":30640,"language":3628,"meta":7,"className":3626,"style":7},"@qml.qnode(dev)\ndef sample_circuit(params, p=1, shots=1000):\n    gammas = params[:p]\n    betas = params[p:]\n    for i in range(n_qubits):\n        qml.Hadamard(wires=i)\n    for layer in range(p):\n        cost_unitary(gammas[layer])\n        mixer_unitary(betas[layer])\n    return qml.sample(wires=range(n_qubits))\n\nsamples = sample_circuit(result.x, shots=1000)\n\n# Count cut values for each bitstring\nfrom collections import Counter\ncut_counts = Counter()\nfor bits in samples:\n    key = ''.join(str(b) for b in bits)\n    cut = sum(1 for u, v in edges if bits[u] != bits[v])\n    cut_counts[(key, cut)] += 1\n\n# Show top 5 most frequent bitstrings\nfor (bits, cut), count in sorted(cut_counts.items(), key=lambda x: -x[1])[:5]:\n    print(f\"{bits}  cut={cut}  frequency={count\u002F10:.1f}%\")\n",[30642],{"type":21,"tag":79,"props":30643,"children":30644},{"__ignoreMap":7},[30645,30656,30697,30712,30727,30750,30769,30792,30799,30806,30834,30841,30874,30881,30889,30910,30927,30948,30997,31053,31069,31076,31084,31147],{"type":21,"tag":322,"props":30646,"children":30647},{"class":324,"line":325},[30648,30652],{"type":21,"tag":322,"props":30649,"children":30650},{"style":4801},[30651],{"type":26,"value":14710},{"type":21,"tag":322,"props":30653,"children":30654},{"style":329},[30655],{"type":26,"value":14715},{"type":21,"tag":322,"props":30657,"children":30658},{"class":324,"line":335},[30659,30663,30668,30672,30676,30680,30685,30689,30693],{"type":21,"tag":322,"props":30660,"children":30661},{"style":3638},[30662],{"type":26,"value":4798},{"type":21,"tag":322,"props":30664,"children":30665},{"style":4801},[30666],{"type":26,"value":30667}," sample_circuit",{"type":21,"tag":322,"props":30669,"children":30670},{"style":329},[30671],{"type":26,"value":29938},{"type":21,"tag":322,"props":30673,"children":30674},{"style":3638},[30675],{"type":26,"value":3678},{"type":21,"tag":322,"props":30677,"children":30678},{"style":348},[30679],{"type":26,"value":3994},{"type":21,"tag":322,"props":30681,"children":30682},{"style":329},[30683],{"type":26,"value":30684},", shots",{"type":21,"tag":322,"props":30686,"children":30687},{"style":3638},[30688],{"type":26,"value":3678},{"type":21,"tag":322,"props":30690,"children":30691},{"style":348},[30692],{"type":26,"value":20897},{"type":21,"tag":322,"props":30694,"children":30695},{"style":329},[30696],{"type":26,"value":4851},{"type":21,"tag":322,"props":30698,"children":30699},{"class":324,"line":344},[30700,30704,30708],{"type":21,"tag":322,"props":30701,"children":30702},{"style":329},[30703],{"type":26,"value":29958},{"type":21,"tag":322,"props":30705,"children":30706},{"style":3638},[30707],{"type":26,"value":3678},{"type":21,"tag":322,"props":30709,"children":30710},{"style":329},[30711],{"type":26,"value":29967},{"type":21,"tag":322,"props":30713,"children":30714},{"class":324,"line":407},[30715,30719,30723],{"type":21,"tag":322,"props":30716,"children":30717},{"style":329},[30718],{"type":26,"value":29975},{"type":21,"tag":322,"props":30720,"children":30721},{"style":3638},[30722],{"type":26,"value":3678},{"type":21,"tag":322,"props":30724,"children":30725},{"style":329},[30726],{"type":26,"value":29984},{"type":21,"tag":322,"props":30728,"children":30729},{"class":324,"line":431},[30730,30734,30738,30742,30746],{"type":21,"tag":322,"props":30731,"children":30732},{"style":3638},[30733],{"type":26,"value":5177},{"type":21,"tag":322,"props":30735,"children":30736},{"style":329},[30737],{"type":26,"value":5182},{"type":21,"tag":322,"props":30739,"children":30740},{"style":3638},[30741],{"type":26,"value":3941},{"type":21,"tag":322,"props":30743,"children":30744},{"style":348},[30745],{"type":26,"value":3946},{"type":21,"tag":322,"props":30747,"children":30748},{"style":329},[30749],{"type":26,"value":27016},{"type":21,"tag":322,"props":30751,"children":30752},{"class":324,"line":453},[30753,30757,30761,30765],{"type":21,"tag":322,"props":30754,"children":30755},{"style":329},[30756],{"type":26,"value":30023},{"type":21,"tag":322,"props":30758,"children":30759},{"style":3703},[30760],{"type":26,"value":14686},{"type":21,"tag":322,"props":30762,"children":30763},{"style":3638},[30764],{"type":26,"value":3678},{"type":21,"tag":322,"props":30766,"children":30767},{"style":329},[30768],{"type":26,"value":27037},{"type":21,"tag":322,"props":30770,"children":30771},{"class":324,"line":474},[30772,30776,30780,30784,30788],{"type":21,"tag":322,"props":30773,"children":30774},{"style":3638},[30775],{"type":26,"value":5177},{"type":21,"tag":322,"props":30777,"children":30778},{"style":329},[30779],{"type":26,"value":30055},{"type":21,"tag":322,"props":30781,"children":30782},{"style":3638},[30783],{"type":26,"value":3941},{"type":21,"tag":322,"props":30785,"children":30786},{"style":348},[30787],{"type":26,"value":3946},{"type":21,"tag":322,"props":30789,"children":30790},{"style":329},[30791],{"type":26,"value":30068},{"type":21,"tag":322,"props":30793,"children":30794},{"class":324,"line":496},[30795],{"type":21,"tag":322,"props":30796,"children":30797},{"style":329},[30798],{"type":26,"value":30076},{"type":21,"tag":322,"props":30800,"children":30801},{"class":324,"line":16},[30802],{"type":21,"tag":322,"props":30803,"children":30804},{"style":329},[30805],{"type":26,"value":30084},{"type":21,"tag":322,"props":30807,"children":30808},{"class":324,"line":3884},[30809,30813,30818,30822,30826,30830],{"type":21,"tag":322,"props":30810,"children":30811},{"style":3638},[30812],{"type":26,"value":5023},{"type":21,"tag":322,"props":30814,"children":30815},{"style":329},[30816],{"type":26,"value":30817}," qml.sample(",{"type":21,"tag":322,"props":30819,"children":30820},{"style":3703},[30821],{"type":26,"value":14686},{"type":21,"tag":322,"props":30823,"children":30824},{"style":3638},[30825],{"type":26,"value":3678},{"type":21,"tag":322,"props":30827,"children":30828},{"style":348},[30829],{"type":26,"value":5379},{"type":21,"tag":322,"props":30831,"children":30832},{"style":329},[30833],{"type":26,"value":14823},{"type":21,"tag":322,"props":30835,"children":30836},{"class":324,"line":3894},[30837],{"type":21,"tag":322,"props":30838,"children":30839},{"emptyLinePlaceholder":3662},[30840],{"type":26,"value":3665},{"type":21,"tag":322,"props":30842,"children":30843},{"class":324,"line":1034},[30844,30849,30853,30858,30862,30866,30870],{"type":21,"tag":322,"props":30845,"children":30846},{"style":329},[30847],{"type":26,"value":30848},"samples ",{"type":21,"tag":322,"props":30850,"children":30851},{"style":3638},[30852],{"type":26,"value":3678},{"type":21,"tag":322,"props":30854,"children":30855},{"style":329},[30856],{"type":26,"value":30857}," sample_circuit(result.x, ",{"type":21,"tag":322,"props":30859,"children":30860},{"style":3703},[30861],{"type":26,"value":4560},{"type":21,"tag":322,"props":30863,"children":30864},{"style":3638},[30865],{"type":26,"value":3678},{"type":21,"tag":322,"props":30867,"children":30868},{"style":348},[30869],{"type":26,"value":20897},{"type":21,"tag":322,"props":30871,"children":30872},{"style":329},[30873],{"type":26,"value":3738},{"type":21,"tag":322,"props":30875,"children":30876},{"class":324,"line":3954},[30877],{"type":21,"tag":322,"props":30878,"children":30879},{"emptyLinePlaceholder":3662},[30880],{"type":26,"value":3665},{"type":21,"tag":322,"props":30882,"children":30883},{"class":324,"line":4002},[30884],{"type":21,"tag":322,"props":30885,"children":30886},{"style":3888},[30887],{"type":26,"value":30888},"# Count cut values for each bitstring\n",{"type":21,"tag":322,"props":30890,"children":30891},{"class":324,"line":4044},[30892,30896,30901,30905],{"type":21,"tag":322,"props":30893,"children":30894},{"style":3638},[30895],{"type":26,"value":3641},{"type":21,"tag":322,"props":30897,"children":30898},{"style":329},[30899],{"type":26,"value":30900}," collections ",{"type":21,"tag":322,"props":30902,"children":30903},{"style":3638},[30904],{"type":26,"value":3651},{"type":21,"tag":322,"props":30906,"children":30907},{"style":329},[30908],{"type":26,"value":30909}," Counter\n",{"type":21,"tag":322,"props":30911,"children":30912},{"class":324,"line":5068},[30913,30918,30922],{"type":21,"tag":322,"props":30914,"children":30915},{"style":329},[30916],{"type":26,"value":30917},"cut_counts ",{"type":21,"tag":322,"props":30919,"children":30920},{"style":3638},[30921],{"type":26,"value":3678},{"type":21,"tag":322,"props":30923,"children":30924},{"style":329},[30925],{"type":26,"value":30926}," Counter()\n",{"type":21,"tag":322,"props":30928,"children":30929},{"class":324,"line":5087},[30930,30934,30939,30943],{"type":21,"tag":322,"props":30931,"children":30932},{"style":3638},[30933],{"type":26,"value":3931},{"type":21,"tag":322,"props":30935,"children":30936},{"style":329},[30937],{"type":26,"value":30938}," bits ",{"type":21,"tag":322,"props":30940,"children":30941},{"style":3638},[30942],{"type":26,"value":3941},{"type":21,"tag":322,"props":30944,"children":30945},{"style":329},[30946],{"type":26,"value":30947}," samples:\n",{"type":21,"tag":322,"props":30949,"children":30950},{"class":324,"line":5095},[30951,30956,30960,30965,30970,30975,30980,30984,30988,30992],{"type":21,"tag":322,"props":30952,"children":30953},{"style":329},[30954],{"type":26,"value":30955},"    key ",{"type":21,"tag":322,"props":30957,"children":30958},{"style":3638},[30959],{"type":26,"value":3678},{"type":21,"tag":322,"props":30961,"children":30962},{"style":359},[30963],{"type":26,"value":30964}," ''",{"type":21,"tag":322,"props":30966,"children":30967},{"style":329},[30968],{"type":26,"value":30969},".join(",{"type":21,"tag":322,"props":30971,"children":30972},{"style":348},[30973],{"type":26,"value":30974},"str",{"type":21,"tag":322,"props":30976,"children":30977},{"style":329},[30978],{"type":26,"value":30979},"(b) ",{"type":21,"tag":322,"props":30981,"children":30982},{"style":3638},[30983],{"type":26,"value":3931},{"type":21,"tag":322,"props":30985,"children":30986},{"style":329},[30987],{"type":26,"value":5341},{"type":21,"tag":322,"props":30989,"children":30990},{"style":3638},[30991],{"type":26,"value":3941},{"type":21,"tag":322,"props":30993,"children":30994},{"style":329},[30995],{"type":26,"value":30996}," bits)\n",{"type":21,"tag":322,"props":30998,"children":30999},{"class":324,"line":5126},[31000,31005,31009,31013,31017,31021,31025,31029,31033,31037,31041,31045,31049],{"type":21,"tag":322,"props":31001,"children":31002},{"style":329},[31003],{"type":26,"value":31004},"    cut ",{"type":21,"tag":322,"props":31006,"children":31007},{"style":3638},[31008],{"type":26,"value":3678},{"type":21,"tag":322,"props":31010,"children":31011},{"style":348},[31012],{"type":26,"value":6757},{"type":21,"tag":322,"props":31014,"children":31015},{"style":329},[31016],{"type":26,"value":3758},{"type":21,"tag":322,"props":31018,"children":31019},{"style":348},[31020],{"type":26,"value":3994},{"type":21,"tag":322,"props":31022,"children":31023},{"style":3638},[31024],{"type":26,"value":30292},{"type":21,"tag":322,"props":31026,"children":31027},{"style":329},[31028],{"type":26,"value":29733},{"type":21,"tag":322,"props":31030,"children":31031},{"style":3638},[31032],{"type":26,"value":3941},{"type":21,"tag":322,"props":31034,"children":31035},{"style":329},[31036],{"type":26,"value":30305},{"type":21,"tag":322,"props":31038,"children":31039},{"style":3638},[31040],{"type":26,"value":30310},{"type":21,"tag":322,"props":31042,"children":31043},{"style":329},[31044],{"type":26,"value":30315},{"type":21,"tag":322,"props":31046,"children":31047},{"style":3638},[31048],{"type":26,"value":30320},{"type":21,"tag":322,"props":31050,"children":31051},{"style":329},[31052],{"type":26,"value":30325},{"type":21,"tag":322,"props":31054,"children":31055},{"class":324,"line":5171},[31056,31061,31065],{"type":21,"tag":322,"props":31057,"children":31058},{"style":329},[31059],{"type":26,"value":31060},"    cut_counts[(key, cut)] ",{"type":21,"tag":322,"props":31062,"children":31063},{"style":3638},[31064],{"type":26,"value":30338},{"type":21,"tag":322,"props":31066,"children":31067},{"style":348},[31068],{"type":26,"value":13231},{"type":21,"tag":322,"props":31070,"children":31071},{"class":324,"line":5210},[31072],{"type":21,"tag":322,"props":31073,"children":31074},{"emptyLinePlaceholder":3662},[31075],{"type":26,"value":3665},{"type":21,"tag":322,"props":31077,"children":31078},{"class":324,"line":5228},[31079],{"type":21,"tag":322,"props":31080,"children":31081},{"style":3888},[31082],{"type":26,"value":31083},"# Show top 5 most frequent bitstrings\n",{"type":21,"tag":322,"props":31085,"children":31086},{"class":324,"line":5302},[31087,31091,31096,31100,31104,31109,31113,31117,31122,31126,31131,31135,31139,31143],{"type":21,"tag":322,"props":31088,"children":31089},{"style":3638},[31090],{"type":26,"value":3931},{"type":21,"tag":322,"props":31092,"children":31093},{"style":329},[31094],{"type":26,"value":31095}," (bits, cut), count ",{"type":21,"tag":322,"props":31097,"children":31098},{"style":3638},[31099],{"type":26,"value":3941},{"type":21,"tag":322,"props":31101,"children":31102},{"style":348},[31103],{"type":26,"value":17576},{"type":21,"tag":322,"props":31105,"children":31106},{"style":329},[31107],{"type":26,"value":31108},"(cut_counts.items(), ",{"type":21,"tag":322,"props":31110,"children":31111},{"style":3703},[31112],{"type":26,"value":3979},{"type":21,"tag":322,"props":31114,"children":31115},{"style":3638},[31116],{"type":26,"value":3984},{"type":21,"tag":322,"props":31118,"children":31119},{"style":329},[31120],{"type":26,"value":31121}," x: ",{"type":21,"tag":322,"props":31123,"children":31124},{"style":3638},[31125],{"type":26,"value":5285},{"type":21,"tag":322,"props":31127,"children":31128},{"style":329},[31129],{"type":26,"value":31130},"x[",{"type":21,"tag":322,"props":31132,"children":31133},{"style":348},[31134],{"type":26,"value":3994},{"type":21,"tag":322,"props":31136,"children":31137},{"style":329},[31138],{"type":26,"value":18015},{"type":21,"tag":322,"props":31140,"children":31141},{"style":348},[31142],{"type":26,"value":7119},{"type":21,"tag":322,"props":31144,"children":31145},{"style":329},[31146],{"type":26,"value":18024},{"type":21,"tag":322,"props":31148,"children":31149},{"class":324,"line":5330},[31150,31154,31158,31162,31166,31170,31174,31178,31183,31187,31192,31196,31201,31205,31210,31214,31218,31223,31227,31232],{"type":21,"tag":322,"props":31151,"children":31152},{"style":348},[31153],{"type":26,"value":17663},{"type":21,"tag":322,"props":31155,"children":31156},{"style":329},[31157],{"type":26,"value":3758},{"type":21,"tag":322,"props":31159,"children":31160},{"style":3638},[31161],{"type":26,"value":3763},{"type":21,"tag":322,"props":31163,"children":31164},{"style":359},[31165],{"type":26,"value":3788},{"type":21,"tag":322,"props":31167,"children":31168},{"style":348},[31169],{"type":26,"value":3773},{"type":21,"tag":322,"props":31171,"children":31172},{"style":329},[31173],{"type":26,"value":17684},{"type":21,"tag":322,"props":31175,"children":31176},{"style":348},[31177],{"type":26,"value":3783},{"type":21,"tag":322,"props":31179,"children":31180},{"style":359},[31181],{"type":26,"value":31182},"  cut=",{"type":21,"tag":322,"props":31184,"children":31185},{"style":348},[31186],{"type":26,"value":3773},{"type":21,"tag":322,"props":31188,"children":31189},{"style":329},[31190],{"type":26,"value":31191},"cut",{"type":21,"tag":322,"props":31193,"children":31194},{"style":348},[31195],{"type":26,"value":3783},{"type":21,"tag":322,"props":31197,"children":31198},{"style":359},[31199],{"type":26,"value":31200},"  frequency=",{"type":21,"tag":322,"props":31202,"children":31203},{"style":348},[31204],{"type":26,"value":3773},{"type":21,"tag":322,"props":31206,"children":31207},{"style":329},[31208],{"type":26,"value":31209},"count",{"type":21,"tag":322,"props":31211,"children":31212},{"style":3638},[31213],{"type":26,"value":174},{"type":21,"tag":322,"props":31215,"children":31216},{"style":348},[31217],{"type":26,"value":20955},{"type":21,"tag":322,"props":31219,"children":31220},{"style":3638},[31221],{"type":26,"value":31222},":.1f",{"type":21,"tag":322,"props":31224,"children":31225},{"style":348},[31226],{"type":26,"value":3783},{"type":21,"tag":322,"props":31228,"children":31229},{"style":359},[31230],{"type":26,"value":31231},"%\"",{"type":21,"tag":322,"props":31233,"children":31234},{"style":329},[31235],{"type":26,"value":3738},{"type":21,"tag":22,"props":31237,"children":31238},{},[31239],{"type":26,"value":31240},"期待される出力：",{"type":21,"tag":312,"props":31242,"children":31244},{"code":31243},"0101  cut=4  frequency=28.3%\n1010  cut=4  frequency=27.1%\n0110  cut=3  frequency=8.4%\n...\n",[31245],{"type":21,"tag":79,"props":31246,"children":31247},{"__ignoreMap":7},[31248],{"type":26,"value":31243},{"type":21,"tag":22,"props":31250,"children":31251},{},[31252,31254,31259,31260,31265],{"type":26,"value":31253},"最適解（",{"type":21,"tag":79,"props":31255,"children":31257},{"className":31256},[],[31258],{"type":26,"value":29378},{"type":26,"value":20934},{"type":21,"tag":79,"props":31261,"children":31263},{"className":31262},[],[31264],{"type":26,"value":17869},{"type":26,"value":31266},"）が支配的になります — QAOAがそれらの確率を増幅したのです。",{"type":21,"tag":59,"props":31268,"children":31270},{"id":31269},"同じ問題をhlquantumで",[31271],{"type":26,"value":31272},"同じ問題をHLQuantumで",{"type":21,"tag":22,"props":31274,"children":31275},{},[31276],{"type":26,"value":31277},"HLQuantumの組み込みQAOAを使えば、回路の構築を完全に省略できます：",{"type":21,"tag":312,"props":31279,"children":31281},{"code":31280,"language":3628,"meta":7,"className":3626,"style":7},"import hlquantum as hlq\n\ngraph = [(0, 1), (1, 2), (2, 3), (3, 0)]\n\nresult = hlq.algorithms.qaoa(\n    graph=graph,\n    problem='max_cut',\n    p=2,\n    backend='pennylane',\n    shots=2000,\n)\n\nprint(result.best_solution)   # '0101' or '1010'\nprint(result.cut_value)       # 4\nprint(result.approximation_ratio)  # ~0.96 for p=2\n",[31282],{"type":21,"tag":79,"props":31283,"children":31284},{"__ignoreMap":7},[31285,31304,31311,31391,31398,31414,31431,31452,31472,31491,31511,31518,31525,31542,31559],{"type":21,"tag":322,"props":31286,"children":31287},{"class":324,"line":325},[31288,31292,31296,31300],{"type":21,"tag":322,"props":31289,"children":31290},{"style":3638},[31291],{"type":26,"value":3651},{"type":21,"tag":322,"props":31293,"children":31294},{"style":329},[31295],{"type":26,"value":19732},{"type":21,"tag":322,"props":31297,"children":31298},{"style":3638},[31299],{"type":26,"value":4715},{"type":21,"tag":322,"props":31301,"children":31302},{"style":329},[31303],{"type":26,"value":19741},{"type":21,"tag":322,"props":31305,"children":31306},{"class":324,"line":335},[31307],{"type":21,"tag":322,"props":31308,"children":31309},{"emptyLinePlaceholder":3662},[31310],{"type":26,"value":3665},{"type":21,"tag":322,"props":31312,"children":31313},{"class":324,"line":344},[31314,31319,31323,31327,31331,31335,31339,31343,31347,31351,31355,31359,31363,31367,31371,31375,31379,31383,31387],{"type":21,"tag":322,"props":31315,"children":31316},{"style":329},[31317],{"type":26,"value":31318},"graph ",{"type":21,"tag":322,"props":31320,"children":31321},{"style":3638},[31322],{"type":26,"value":3678},{"type":21,"tag":322,"props":31324,"children":31325},{"style":329},[31326],{"type":26,"value":29566},{"type":21,"tag":322,"props":31328,"children":31329},{"style":348},[31330],{"type":26,"value":5605},{"type":21,"tag":322,"props":31332,"children":31333},{"style":329},[31334],{"type":26,"value":367},{"type":21,"tag":322,"props":31336,"children":31337},{"style":348},[31338],{"type":26,"value":3994},{"type":21,"tag":322,"props":31340,"children":31341},{"style":329},[31342],{"type":26,"value":29583},{"type":21,"tag":322,"props":31344,"children":31345},{"style":348},[31346],{"type":26,"value":3994},{"type":21,"tag":322,"props":31348,"children":31349},{"style":329},[31350],{"type":26,"value":367},{"type":21,"tag":322,"props":31352,"children":31353},{"style":348},[31354],{"type":26,"value":423},{"type":21,"tag":322,"props":31356,"children":31357},{"style":329},[31358],{"type":26,"value":29583},{"type":21,"tag":322,"props":31360,"children":31361},{"style":348},[31362],{"type":26,"value":423},{"type":21,"tag":322,"props":31364,"children":31365},{"style":329},[31366],{"type":26,"value":367},{"type":21,"tag":322,"props":31368,"children":31369},{"style":348},[31370],{"type":26,"value":6948},{"type":21,"tag":322,"props":31372,"children":31373},{"style":329},[31374],{"type":26,"value":29583},{"type":21,"tag":322,"props":31376,"children":31377},{"style":348},[31378],{"type":26,"value":6948},{"type":21,"tag":322,"props":31380,"children":31381},{"style":329},[31382],{"type":26,"value":367},{"type":21,"tag":322,"props":31384,"children":31385},{"style":348},[31386],{"type":26,"value":5605},{"type":21,"tag":322,"props":31388,"children":31389},{"style":329},[31390],{"type":26,"value":29632},{"type":21,"tag":322,"props":31392,"children":31393},{"class":324,"line":407},[31394],{"type":21,"tag":322,"props":31395,"children":31396},{"emptyLinePlaceholder":3662},[31397],{"type":26,"value":3665},{"type":21,"tag":322,"props":31399,"children":31400},{"class":324,"line":431},[31401,31405,31409],{"type":21,"tag":322,"props":31402,"children":31403},{"style":329},[31404],{"type":26,"value":19764},{"type":21,"tag":322,"props":31406,"children":31407},{"style":3638},[31408],{"type":26,"value":3678},{"type":21,"tag":322,"props":31410,"children":31411},{"style":329},[31412],{"type":26,"value":31413}," hlq.algorithms.qaoa(\n",{"type":21,"tag":322,"props":31415,"children":31416},{"class":324,"line":453},[31417,31422,31426],{"type":21,"tag":322,"props":31418,"children":31419},{"style":3703},[31420],{"type":26,"value":31421},"    graph",{"type":21,"tag":322,"props":31423,"children":31424},{"style":3638},[31425],{"type":26,"value":3678},{"type":21,"tag":322,"props":31427,"children":31428},{"style":329},[31429],{"type":26,"value":31430},"graph,\n",{"type":21,"tag":322,"props":31432,"children":31433},{"class":324,"line":474},[31434,31439,31443,31448],{"type":21,"tag":322,"props":31435,"children":31436},{"style":3703},[31437],{"type":26,"value":31438},"    problem",{"type":21,"tag":322,"props":31440,"children":31441},{"style":3638},[31442],{"type":26,"value":3678},{"type":21,"tag":322,"props":31444,"children":31445},{"style":359},[31446],{"type":26,"value":31447},"'max_cut'",{"type":21,"tag":322,"props":31449,"children":31450},{"style":329},[31451],{"type":26,"value":428},{"type":21,"tag":322,"props":31453,"children":31454},{"class":324,"line":496},[31455,31460,31464,31468],{"type":21,"tag":322,"props":31456,"children":31457},{"style":3703},[31458],{"type":26,"value":31459},"    p",{"type":21,"tag":322,"props":31461,"children":31462},{"style":3638},[31463],{"type":26,"value":3678},{"type":21,"tag":322,"props":31465,"children":31466},{"style":348},[31467],{"type":26,"value":423},{"type":21,"tag":322,"props":31469,"children":31470},{"style":329},[31471],{"type":26,"value":428},{"type":21,"tag":322,"props":31473,"children":31474},{"class":324,"line":16},[31475,31479,31483,31487],{"type":21,"tag":322,"props":31476,"children":31477},{"style":3703},[31478],{"type":26,"value":23300},{"type":21,"tag":322,"props":31480,"children":31481},{"style":3638},[31482],{"type":26,"value":3678},{"type":21,"tag":322,"props":31484,"children":31485},{"style":359},[31486],{"type":26,"value":22610},{"type":21,"tag":322,"props":31488,"children":31489},{"style":329},[31490],{"type":26,"value":428},{"type":21,"tag":322,"props":31492,"children":31493},{"class":324,"line":3884},[31494,31498,31502,31507],{"type":21,"tag":322,"props":31495,"children":31496},{"style":3703},[31497],{"type":26,"value":29245},{"type":21,"tag":322,"props":31499,"children":31500},{"style":3638},[31501],{"type":26,"value":3678},{"type":21,"tag":322,"props":31503,"children":31504},{"style":348},[31505],{"type":26,"value":31506},"2000",{"type":21,"tag":322,"props":31508,"children":31509},{"style":329},[31510],{"type":26,"value":428},{"type":21,"tag":322,"props":31512,"children":31513},{"class":324,"line":3894},[31514],{"type":21,"tag":322,"props":31515,"children":31516},{"style":329},[31517],{"type":26,"value":3738},{"type":21,"tag":322,"props":31519,"children":31520},{"class":324,"line":1034},[31521],{"type":21,"tag":322,"props":31522,"children":31523},{"emptyLinePlaceholder":3662},[31524],{"type":26,"value":3665},{"type":21,"tag":322,"props":31526,"children":31527},{"class":324,"line":3954},[31528,31532,31537],{"type":21,"tag":322,"props":31529,"children":31530},{"style":348},[31531],{"type":26,"value":3753},{"type":21,"tag":322,"props":31533,"children":31534},{"style":329},[31535],{"type":26,"value":31536},"(result.best_solution)   ",{"type":21,"tag":322,"props":31538,"children":31539},{"style":3888},[31540],{"type":26,"value":31541},"# '0101' or '1010'\n",{"type":21,"tag":322,"props":31543,"children":31544},{"class":324,"line":4002},[31545,31549,31554],{"type":21,"tag":322,"props":31546,"children":31547},{"style":348},[31548],{"type":26,"value":3753},{"type":21,"tag":322,"props":31550,"children":31551},{"style":329},[31552],{"type":26,"value":31553},"(result.cut_value)       ",{"type":21,"tag":322,"props":31555,"children":31556},{"style":3888},[31557],{"type":26,"value":31558},"# 4\n",{"type":21,"tag":322,"props":31560,"children":31561},{"class":324,"line":4044},[31562,31566,31571],{"type":21,"tag":322,"props":31563,"children":31564},{"style":348},[31565],{"type":26,"value":3753},{"type":21,"tag":322,"props":31567,"children":31568},{"style":329},[31569],{"type":26,"value":31570},"(result.approximation_ratio)  ",{"type":21,"tag":322,"props":31572,"children":31573},{"style":3888},[31574],{"type":26,"value":31575},"# ~0.96 for p=2\n",{"type":21,"tag":22,"props":31577,"children":31578},{},[31579],{"type":26,"value":31580},"単一のフラグでバックエンドを切り替えられます：",{"type":21,"tag":312,"props":31582,"children":31584},{"code":31583,"language":3628,"meta":7,"className":3626,"style":7},"# Run on Qiskit Aer\nresult_qiskit = hlq.algorithms.qaoa(graph=graph, problem='max_cut', p=2, backend='qiskit')\n\n# Run on real IonQ hardware (if you have credits)\nresult_ionq = hlq.algorithms.qaoa(graph=graph, problem='max_cut', p=2,\n                                   backend='ionq', device='aria-1')\n",[31585],{"type":21,"tag":79,"props":31586,"children":31587},{"__ignoreMap":7},[31588,31596,31675,31682,31690,31750],{"type":21,"tag":322,"props":31589,"children":31590},{"class":324,"line":325},[31591],{"type":21,"tag":322,"props":31592,"children":31593},{"style":3888},[31594],{"type":26,"value":31595},"# Run on Qiskit Aer\n",{"type":21,"tag":322,"props":31597,"children":31598},{"class":324,"line":335},[31599,31603,31607,31612,31617,31621,31626,31631,31635,31639,31643,31647,31651,31655,31659,31663,31667,31671],{"type":21,"tag":322,"props":31600,"children":31601},{"style":329},[31602],{"type":26,"value":22051},{"type":21,"tag":322,"props":31604,"children":31605},{"style":3638},[31606],{"type":26,"value":3678},{"type":21,"tag":322,"props":31608,"children":31609},{"style":329},[31610],{"type":26,"value":31611}," hlq.algorithms.qaoa(",{"type":21,"tag":322,"props":31613,"children":31614},{"style":3703},[31615],{"type":26,"value":31616},"graph",{"type":21,"tag":322,"props":31618,"children":31619},{"style":3638},[31620],{"type":26,"value":3678},{"type":21,"tag":322,"props":31622,"children":31623},{"style":329},[31624],{"type":26,"value":31625},"graph, ",{"type":21,"tag":322,"props":31627,"children":31628},{"style":3703},[31629],{"type":26,"value":31630},"problem",{"type":21,"tag":322,"props":31632,"children":31633},{"style":3638},[31634],{"type":26,"value":3678},{"type":21,"tag":322,"props":31636,"children":31637},{"style":359},[31638],{"type":26,"value":31447},{"type":21,"tag":322,"props":31640,"children":31641},{"style":329},[31642],{"type":26,"value":367},{"type":21,"tag":322,"props":31644,"children":31645},{"style":3703},[31646],{"type":26,"value":22},{"type":21,"tag":322,"props":31648,"children":31649},{"style":3638},[31650],{"type":26,"value":3678},{"type":21,"tag":322,"props":31652,"children":31653},{"style":348},[31654],{"type":26,"value":423},{"type":21,"tag":322,"props":31656,"children":31657},{"style":329},[31658],{"type":26,"value":367},{"type":21,"tag":322,"props":31660,"children":31661},{"style":3703},[31662],{"type":26,"value":18908},{"type":21,"tag":322,"props":31664,"children":31665},{"style":3638},[31666],{"type":26,"value":3678},{"type":21,"tag":322,"props":31668,"children":31669},{"style":359},[31670],{"type":26,"value":19822},{"type":21,"tag":322,"props":31672,"children":31673},{"style":329},[31674],{"type":26,"value":3738},{"type":21,"tag":322,"props":31676,"children":31677},{"class":324,"line":344},[31678],{"type":21,"tag":322,"props":31679,"children":31680},{"emptyLinePlaceholder":3662},[31681],{"type":26,"value":3665},{"type":21,"tag":322,"props":31683,"children":31684},{"class":324,"line":407},[31685],{"type":21,"tag":322,"props":31686,"children":31687},{"style":3888},[31688],{"type":26,"value":31689},"# Run on real IonQ hardware (if you have credits)\n",{"type":21,"tag":322,"props":31691,"children":31692},{"class":324,"line":431},[31693,31698,31702,31706,31710,31714,31718,31722,31726,31730,31734,31738,31742,31746],{"type":21,"tag":322,"props":31694,"children":31695},{"style":329},[31696],{"type":26,"value":31697},"result_ionq ",{"type":21,"tag":322,"props":31699,"children":31700},{"style":3638},[31701],{"type":26,"value":3678},{"type":21,"tag":322,"props":31703,"children":31704},{"style":329},[31705],{"type":26,"value":31611},{"type":21,"tag":322,"props":31707,"children":31708},{"style":3703},[31709],{"type":26,"value":31616},{"type":21,"tag":322,"props":31711,"children":31712},{"style":3638},[31713],{"type":26,"value":3678},{"type":21,"tag":322,"props":31715,"children":31716},{"style":329},[31717],{"type":26,"value":31625},{"type":21,"tag":322,"props":31719,"children":31720},{"style":3703},[31721],{"type":26,"value":31630},{"type":21,"tag":322,"props":31723,"children":31724},{"style":3638},[31725],{"type":26,"value":3678},{"type":21,"tag":322,"props":31727,"children":31728},{"style":359},[31729],{"type":26,"value":31447},{"type":21,"tag":322,"props":31731,"children":31732},{"style":329},[31733],{"type":26,"value":367},{"type":21,"tag":322,"props":31735,"children":31736},{"style":3703},[31737],{"type":26,"value":22},{"type":21,"tag":322,"props":31739,"children":31740},{"style":3638},[31741],{"type":26,"value":3678},{"type":21,"tag":322,"props":31743,"children":31744},{"style":348},[31745],{"type":26,"value":423},{"type":21,"tag":322,"props":31747,"children":31748},{"style":329},[31749],{"type":26,"value":428},{"type":21,"tag":322,"props":31751,"children":31752},{"class":324,"line":453},[31753,31758,31762,31766,31770,31774,31778,31782],{"type":21,"tag":322,"props":31754,"children":31755},{"style":3703},[31756],{"type":26,"value":31757},"                                   backend",{"type":21,"tag":322,"props":31759,"children":31760},{"style":3638},[31761],{"type":26,"value":3678},{"type":21,"tag":322,"props":31763,"children":31764},{"style":359},[31765],{"type":26,"value":22714},{"type":21,"tag":322,"props":31767,"children":31768},{"style":329},[31769],{"type":26,"value":367},{"type":21,"tag":322,"props":31771,"children":31772},{"style":3703},[31773],{"type":26,"value":22723},{"type":21,"tag":322,"props":31775,"children":31776},{"style":3638},[31777],{"type":26,"value":3678},{"type":21,"tag":322,"props":31779,"children":31780},{"style":359},[31781],{"type":26,"value":22732},{"type":21,"tag":322,"props":31783,"children":31784},{"style":329},[31785],{"type":26,"value":3738},{"type":21,"tag":59,"props":31787,"children":31789},{"id":31788},"より良い結果のためにpを増やす",[31790],{"type":26,"value":31788},{"type":21,"tag":22,"props":31792,"children":31793},{},[31794,31796,31801,31803,31809],{"type":26,"value":31795},"QAOAの近似比は深さ ",{"type":21,"tag":79,"props":31797,"children":31799},{"className":31798},[],[31800],{"type":26,"value":22},{"type":26,"value":31802}," とともに向上します。",{"type":21,"tag":79,"props":31804,"children":31806},{"className":31805},[],[31807],{"type":26,"value":31808},"p→∞",{"type":26,"value":31810}," では、QAOAは問題を厳密に解きます。実際には：",{"type":21,"tag":4503,"props":31812,"children":31813},{},[31814,31834],{"type":21,"tag":4507,"props":31815,"children":31816},{},[31817],{"type":21,"tag":4511,"props":31818,"children":31819},{},[31820,31824,31829],{"type":21,"tag":4515,"props":31821,"children":31822},{},[31823],{"type":26,"value":22},{"type":21,"tag":4515,"props":31825,"children":31826},{},[31827],{"type":26,"value":31828},"Max-Cutでの典型的な近似比",{"type":21,"tag":4515,"props":31830,"children":31831},{},[31832],{"type":26,"value":31833},"必要なゲート数",{"type":21,"tag":4529,"props":31835,"children":31836},{},[31837,31854,31871,31888],{"type":21,"tag":4511,"props":31838,"children":31839},{},[31840,31844,31849],{"type":21,"tag":4536,"props":31841,"children":31842},{},[31843],{"type":26,"value":3994},{"type":21,"tag":4536,"props":31845,"children":31846},{},[31847],{"type":26,"value":31848},"~0.75",{"type":21,"tag":4536,"props":31850,"children":31851},{},[31852],{"type":26,"value":31853},"2×edges + 4n",{"type":21,"tag":4511,"props":31855,"children":31856},{},[31857,31861,31866],{"type":21,"tag":4536,"props":31858,"children":31859},{},[31860],{"type":26,"value":423},{"type":21,"tag":4536,"props":31862,"children":31863},{},[31864],{"type":26,"value":31865},"~0.86",{"type":21,"tag":4536,"props":31867,"children":31868},{},[31869],{"type":26,"value":31870},"4×edges + 8n",{"type":21,"tag":4511,"props":31872,"children":31873},{},[31874,31878,31883],{"type":21,"tag":4536,"props":31875,"children":31876},{},[31877],{"type":26,"value":6948},{"type":21,"tag":4536,"props":31879,"children":31880},{},[31881],{"type":26,"value":31882},"~0.91",{"type":21,"tag":4536,"props":31884,"children":31885},{},[31886],{"type":26,"value":31887},"6×edges + 12n",{"type":21,"tag":4511,"props":31889,"children":31890},{},[31891,31895,31900],{"type":21,"tag":4536,"props":31892,"children":31893},{},[31894],{"type":26,"value":7119},{"type":21,"tag":4536,"props":31896,"children":31897},{},[31898],{"type":26,"value":31899},"~0.95",{"type":21,"tag":4536,"props":31901,"children":31902},{},[31903],{"type":26,"value":31904},"10×edges + 20n",{"type":21,"tag":22,"props":31906,"children":31907},{},[31908,31910,31916,31918,31923,31924,31930],{"type":26,"value":31909},"トレードオフとして、深い回路は実ハードウェア上でより多くのノイズを蓄積します。シミュレータ上では ",{"type":21,"tag":79,"props":31911,"children":31913},{"className":31912},[],[31914],{"type":26,"value":31915},"p=3",{"type":26,"value":31917}," 以上を使いましょう。NISQ QPU上では、",{"type":21,"tag":79,"props":31919,"children":31921},{"className":31920},[],[31922],{"type":26,"value":19975},{"type":26,"value":19246},{"type":21,"tag":79,"props":31925,"children":31927},{"className":31926},[],[31928],{"type":26,"value":31929},"p=2",{"type":26,"value":31931}," が通常の実用的な限界です。",{"type":21,"tag":59,"props":31933,"children":31935},{"id":31934},"qaoaはいつ使う価値があるのか",[31936],{"type":26,"value":31937},"QAOAはいつ使う価値があるのか",{"type":21,"tag":22,"props":31939,"children":31940},{},[31941],{"type":21,"tag":71,"props":31942,"children":31943},{},[31944],{"type":26,"value":31945},"現在有用な用途：",{"type":21,"tag":3556,"props":31947,"children":31948},{},[31949,31954,31959],{"type":21,"tag":3560,"props":31950,"children":31951},{},[31952],{"type":26,"value":31953},"NISQデバイスのベンチマーク（QAOAは標準的なベンチマーク回路）",{"type":21,"tag":3560,"props":31955,"children":31956},{},[31957],{"type":26,"value":31958},"変分量子アルゴリズムに関する学術研究",{"type":21,"tag":3560,"props":31960,"children":31961},{},[31962],{"type":26,"value":31963},"シミュレータ上の小規模グラフ（20ノード未満）",{"type":21,"tag":22,"props":31965,"children":31966},{},[31967],{"type":21,"tag":71,"props":31968,"children":31969},{},[31970],{"type":26,"value":31971},"まだ有用ではない用途：",{"type":21,"tag":3556,"props":31973,"children":31974},{},[31975,31980],{"type":21,"tag":3560,"props":31976,"children":31977},{},[31978],{"type":26,"value":31979},"実ハードウェア上での大規模な本番最適化 — 古典的ヒューリスティクス（シミュレーテッドアニーリング、タブーサーチ）は実用的な問題において依然としてQAOAを上回る",{"type":21,"tag":3560,"props":31981,"children":31982},{},[31983],{"type":26,"value":31984},"大規模グラフ — 回路の深さはグラフの密度とともに増大する",{"type":21,"tag":22,"props":31986,"children":31987},{},[31988,31993,31995,32000],{"type":21,"tag":71,"props":31989,"children":31990},{},[31991],{"type":26,"value":31992},"その可能性：",{"type":26,"value":31994}," 高深度回路を扱える将来の誤り耐性ハードウェア上では、大きな ",{"type":21,"tag":79,"props":31996,"children":31998},{"className":31997},[],[31999],{"type":26,"value":22},{"type":26,"value":32001}," でのQAOAが密なNP困難最適化に対して真の高速化をもたらしうるかもしれません。理論的基盤は堅固ですが、ハードウェアはまだそこに達していません。",{"type":21,"tag":59,"props":32003,"children":32005},{"id":32004},"次のステップ",[32006],{"type":26,"value":32004},{"type":21,"tag":3556,"props":32008,"children":32009},{},[32010,32023,32036],{"type":21,"tag":3560,"props":32011,"children":32012},{},[32013,32021],{"type":21,"tag":71,"props":32014,"children":32015},{},[32016],{"type":21,"tag":29,"props":32017,"children":32018},{"href":14507},[32019],{"type":26,"value":32020},"PennyLane SDKガイド",{"type":26,"value":32022}," — より多くの変分アルゴリズムのためにPennyLaneをセットアップする",{"type":21,"tag":3560,"props":32024,"children":32025},{},[32026,32034],{"type":21,"tag":71,"props":32027,"children":32028},{},[32029],{"type":21,"tag":29,"props":32030,"children":32031},{"href":893},[32032],{"type":26,"value":32033},"VQEガイド",{"type":26,"value":32035}," — 量子化学のための姉妹アルゴリズム",{"type":21,"tag":3560,"props":32037,"children":32038},{},[32039,32046],{"type":21,"tag":71,"props":32040,"children":32041},{},[32042],{"type":21,"tag":29,"props":32043,"children":32044},{"href":19441},[32045],{"type":26,"value":19444},{"type":26,"value":32047}," — 一つのAPIで任意のバックエンド上でQAOAを実行する",{"type":21,"tag":703,"props":32049,"children":32050},{},[32051],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":32053},[32054,32055,32056,32057,32058,32059,32060,32061],{"id":29346,"depth":335,"text":29349},{"id":29389,"depth":335,"text":29392},{"id":29465,"depth":335,"text":29468},{"id":30630,"depth":335,"text":30630},{"id":31269,"depth":335,"text":31272},{"id":31788,"depth":335,"text":31788},{"id":31934,"depth":335,"text":31937},{"id":32004,"depth":335,"text":32004},"content:ja:blog:qaoa-max-cut-tutorial.md","ja\u002Fblog\u002Fqaoa-max-cut-tutorial.md","ja\u002Fblog\u002Fqaoa-max-cut-tutorial",{"_path":32066,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":32067,"description":32068,"date":32069,"author":11,"tags":32070,"readingTime":474,"body":32072,"_type":715,"_id":34059,"_source":717,"_file":34060,"_stem":34061,"_extension":720},"\u002Fja\u002Fblog\u002Fquantum-simulators-compared","無料の量子シミュレータ比較: Qiskit Aer vs Cirq vs PennyLane vs CUDA-Q","主要な無料量子シミュレータの実践的な比較 — 性能、量子ビット上限、GPU対応、そしてそれぞれをいつ使うべきか。","2026-02-10",[20608,3502,32071],"GPU",{"type":18,"children":32073,"toc":34050},[32074,32079,32084,32267,32273,32282,32335,32505,32515,32521,32537,32864,32884,32907,32955,32965,32971,32980,33217,33227,33245,33255,33261,33270,33446,33456,33462,33473,33686,33696,33702,33713,34041,34046],{"type":21,"tag":22,"props":32075,"children":32076},{},[32077],{"type":26,"value":32078},"量子シミュレータを使えば、ハードウェアのキューを待ったりショットごとに料金を支払ったりすることなく回路をテストできます。しかし、無料で利用できる主要なシミュレータが6つもある中で、どれを使うべきでしょうか。ここでは実践的な整理をお届けします。",{"type":21,"tag":59,"props":32080,"children":32082},{"id":32081},"手っ取り早い答え",[32083],{"type":26,"value":32081},{"type":21,"tag":4503,"props":32085,"children":32086},{},[32087,32112],{"type":21,"tag":4507,"props":32088,"children":32089},{},[32090],{"type":21,"tag":4511,"props":32091,"children":32092},{},[32093,32098,32102,32107],{"type":21,"tag":4515,"props":32094,"children":32095},{},[32096],{"type":26,"value":32097},"シミュレータ",{"type":21,"tag":4515,"props":32099,"children":32100},{},[32101],{"type":26,"value":21229},{"type":21,"tag":4515,"props":32103,"children":32104},{},[32105],{"type":26,"value":32106},"最大量子ビット数（CPU）",{"type":21,"tag":4515,"props":32108,"children":32109},{},[32110],{"type":26,"value":32111},"GPU?",{"type":21,"tag":4529,"props":32113,"children":32114},{},[32115,32141,32167,32193,32218,32243],{"type":21,"tag":4511,"props":32116,"children":32117},{},[32118,32126,32131,32136],{"type":21,"tag":4536,"props":32119,"children":32120},{},[32121],{"type":21,"tag":71,"props":32122,"children":32123},{},[32124],{"type":26,"value":32125},"Qiskit Aer",{"type":21,"tag":4536,"props":32127,"children":32128},{},[32129],{"type":26,"value":32130},"汎用",{"type":21,"tag":4536,"props":32132,"children":32133},{},[32134],{"type":26,"value":32135},"約30（状態ベクトル）",{"type":21,"tag":4536,"props":32137,"children":32138},{},[32139],{"type":26,"value":32140},"✅ AerCuda",{"type":21,"tag":4511,"props":32142,"children":32143},{},[32144,32152,32157,32162],{"type":21,"tag":4536,"props":32145,"children":32146},{},[32147],{"type":21,"tag":71,"props":32148,"children":32149},{},[32150],{"type":26,"value":32151},"Cirq Simulator",{"type":21,"tag":4536,"props":32153,"children":32154},{},[32155],{"type":26,"value":32156},"Cirq回路、密度行列",{"type":21,"tag":4536,"props":32158,"children":32159},{},[32160],{"type":26,"value":32161},"約25",{"type":21,"tag":4536,"props":32163,"children":32164},{},[32165],{"type":26,"value":32166},"❌",{"type":21,"tag":4511,"props":32168,"children":32169},{},[32170,32178,32183,32188],{"type":21,"tag":4536,"props":32171,"children":32172},{},[32173],{"type":21,"tag":71,"props":32174,"children":32175},{},[32176],{"type":26,"value":32177},"PennyLane default.qubit",{"type":21,"tag":4536,"props":32179,"children":32180},{},[32181],{"type":26,"value":32182},"QML、勾配",{"type":21,"tag":4536,"props":32184,"children":32185},{},[32186],{"type":26,"value":32187},"約20",{"type":21,"tag":4536,"props":32189,"children":32190},{},[32191],{"type":26,"value":32192},"✅ lightning.gpu",{"type":21,"tag":4511,"props":32194,"children":32195},{},[32196,32204,32209,32214],{"type":21,"tag":4536,"props":32197,"children":32198},{},[32199],{"type":21,"tag":71,"props":32200,"children":32201},{},[32202],{"type":26,"value":32203},"PennyLane lightning.qubit",{"type":21,"tag":4536,"props":32205,"children":32206},{},[32207],{"type":26,"value":32208},"高速なCPUシミュレーション",{"type":21,"tag":4536,"props":32210,"children":32211},{},[32212],{"type":26,"value":32213},"約30",{"type":21,"tag":4536,"props":32215,"children":32216},{},[32217],{"type":26,"value":32192},{"type":21,"tag":4511,"props":32219,"children":32220},{},[32221,32228,32233,32238],{"type":21,"tag":4536,"props":32222,"children":32223},{},[32224],{"type":21,"tag":71,"props":32225,"children":32226},{},[32227],{"type":26,"value":21053},{"type":21,"tag":4536,"props":32229,"children":32230},{},[32231],{"type":26,"value":32232},"大規模回路、速度",{"type":21,"tag":4536,"props":32234,"children":32235},{},[32236],{"type":26,"value":32237},"シングルGPUで約34",{"type":21,"tag":4536,"props":32239,"children":32240},{},[32241],{"type":26,"value":32242},"✅ ネイティブ",{"type":21,"tag":4511,"props":32244,"children":32245},{},[32246,32254,32259,32263],{"type":21,"tag":4536,"props":32247,"children":32248},{},[32249],{"type":21,"tag":71,"props":32250,"children":32251},{},[32252],{"type":26,"value":32253},"Braket LocalSimulator",{"type":21,"tag":4536,"props":32255,"children":32256},{},[32257],{"type":26,"value":32258},"Braket回路、無料",{"type":21,"tag":4536,"props":32260,"children":32261},{},[32262],{"type":26,"value":32161},{"type":21,"tag":4536,"props":32264,"children":32265},{},[32266],{"type":26,"value":32166},{"type":21,"tag":59,"props":32268,"children":32270},{"id":32269},"qiskit-aer-働き者",[32271],{"type":26,"value":32272},"Qiskit Aer: 働き者",{"type":21,"tag":22,"props":32274,"children":32275},{},[32276,32280],{"type":21,"tag":29,"props":32277,"children":32278},{"href":7528},[32279],{"type":26,"value":32125},{"type":26,"value":32281}," は最も機能が充実した無料シミュレータです。複数のシミュレーション手法をサポートしています:",{"type":21,"tag":3556,"props":32283,"children":32284},{},[32285,32295,32305,32315,32325],{"type":21,"tag":3560,"props":32286,"children":32287},{},[32288,32293],{"type":21,"tag":71,"props":32289,"children":32290},{},[32291],{"type":26,"value":32292},"statevector_simulator",{"type":26,"value":32294},": 約30量子ビットまでの厳密なシミュレーション。メモリは 2ⁿ 個の複素数としてスケールします（30量子ビットには16 GBのRAMが必要）。",{"type":21,"tag":3560,"props":32296,"children":32297},{},[32298,32303],{"type":21,"tag":71,"props":32299,"children":32300},{},[32301],{"type":26,"value":32302},"qasm_simulator",{"type":26,"value":32304},": ノイズモデルに対応したショットベースのサンプリング",{"type":21,"tag":3560,"props":32306,"children":32307},{},[32308,32313],{"type":21,"tag":71,"props":32309,"children":32310},{},[32311],{"type":26,"value":32312},"density_matrix",{"type":26,"value":32314},": 混合状態と開放量子系をシミュレーション",{"type":21,"tag":3560,"props":32316,"children":32317},{},[32318,32323],{"type":21,"tag":71,"props":32319,"children":32320},{},[32321],{"type":26,"value":32322},"mps",{"type":26,"value":32324},": 行列積状態（Matrix Product State） — 限られたエンタングルメントを持つ回路を数百量子ビットまでシミュレーション",{"type":21,"tag":3560,"props":32326,"children":32327},{},[32328,32333],{"type":21,"tag":71,"props":32329,"children":32330},{},[32331],{"type":26,"value":32332},"stabilizer",{"type":26,"value":32334},": Clifford回路のみですが、多項式時間で数千量子ビットまでスケールします",{"type":21,"tag":312,"props":32336,"children":32338},{"className":3626,"code":32337,"language":3628,"meta":7,"style":7},"from qiskit_aer import AerSimulator\n\n# Default: automatic method selection\nsim = AerSimulator()\n\n# Force a specific method\nsim_sv = AerSimulator(method='statevector')\nsim_dm = AerSimulator(method='density_matrix')\nsim_mps = AerSimulator(method='matrix_product_state')\n",[32339],{"type":21,"tag":79,"props":32340,"children":32341},{"__ignoreMap":7},[32342,32361,32368,32376,32391,32398,32406,32439,32472],{"type":21,"tag":322,"props":32343,"children":32344},{"class":324,"line":325},[32345,32349,32353,32357],{"type":21,"tag":322,"props":32346,"children":32347},{"style":3638},[32348],{"type":26,"value":3641},{"type":21,"tag":322,"props":32350,"children":32351},{"style":329},[32352],{"type":26,"value":4753},{"type":21,"tag":322,"props":32354,"children":32355},{"style":3638},[32356],{"type":26,"value":3651},{"type":21,"tag":322,"props":32358,"children":32359},{"style":329},[32360],{"type":26,"value":4762},{"type":21,"tag":322,"props":32362,"children":32363},{"class":324,"line":335},[32364],{"type":21,"tag":322,"props":32365,"children":32366},{"emptyLinePlaceholder":3662},[32367],{"type":26,"value":3665},{"type":21,"tag":322,"props":32369,"children":32370},{"class":324,"line":344},[32371],{"type":21,"tag":322,"props":32372,"children":32373},{"style":3888},[32374],{"type":26,"value":32375},"# Default: automatic method selection\n",{"type":21,"tag":322,"props":32377,"children":32378},{"class":324,"line":407},[32379,32383,32387],{"type":21,"tag":322,"props":32380,"children":32381},{"style":329},[32382],{"type":26,"value":8861},{"type":21,"tag":322,"props":32384,"children":32385},{"style":3638},[32386],{"type":26,"value":3678},{"type":21,"tag":322,"props":32388,"children":32389},{"style":329},[32390],{"type":26,"value":8870},{"type":21,"tag":322,"props":32392,"children":32393},{"class":324,"line":431},[32394],{"type":21,"tag":322,"props":32395,"children":32396},{"emptyLinePlaceholder":3662},[32397],{"type":26,"value":3665},{"type":21,"tag":322,"props":32399,"children":32400},{"class":324,"line":453},[32401],{"type":21,"tag":322,"props":32402,"children":32403},{"style":3888},[32404],{"type":26,"value":32405},"# Force a specific method\n",{"type":21,"tag":322,"props":32407,"children":32408},{"class":324,"line":474},[32409,32414,32418,32422,32426,32430,32435],{"type":21,"tag":322,"props":32410,"children":32411},{"style":329},[32412],{"type":26,"value":32413},"sim_sv ",{"type":21,"tag":322,"props":32415,"children":32416},{"style":3638},[32417],{"type":26,"value":3678},{"type":21,"tag":322,"props":32419,"children":32420},{"style":329},[32421],{"type":26,"value":5051},{"type":21,"tag":322,"props":32423,"children":32424},{"style":3703},[32425],{"type":26,"value":30463},{"type":21,"tag":322,"props":32427,"children":32428},{"style":3638},[32429],{"type":26,"value":3678},{"type":21,"tag":322,"props":32431,"children":32432},{"style":359},[32433],{"type":26,"value":32434},"'statevector'",{"type":21,"tag":322,"props":32436,"children":32437},{"style":329},[32438],{"type":26,"value":3738},{"type":21,"tag":322,"props":32440,"children":32441},{"class":324,"line":496},[32442,32447,32451,32455,32459,32463,32468],{"type":21,"tag":322,"props":32443,"children":32444},{"style":329},[32445],{"type":26,"value":32446},"sim_dm ",{"type":21,"tag":322,"props":32448,"children":32449},{"style":3638},[32450],{"type":26,"value":3678},{"type":21,"tag":322,"props":32452,"children":32453},{"style":329},[32454],{"type":26,"value":5051},{"type":21,"tag":322,"props":32456,"children":32457},{"style":3703},[32458],{"type":26,"value":30463},{"type":21,"tag":322,"props":32460,"children":32461},{"style":3638},[32462],{"type":26,"value":3678},{"type":21,"tag":322,"props":32464,"children":32465},{"style":359},[32466],{"type":26,"value":32467},"'density_matrix'",{"type":21,"tag":322,"props":32469,"children":32470},{"style":329},[32471],{"type":26,"value":3738},{"type":21,"tag":322,"props":32473,"children":32474},{"class":324,"line":16},[32475,32480,32484,32488,32492,32496,32501],{"type":21,"tag":322,"props":32476,"children":32477},{"style":329},[32478],{"type":26,"value":32479},"sim_mps ",{"type":21,"tag":322,"props":32481,"children":32482},{"style":3638},[32483],{"type":26,"value":3678},{"type":21,"tag":322,"props":32485,"children":32486},{"style":329},[32487],{"type":26,"value":5051},{"type":21,"tag":322,"props":32489,"children":32490},{"style":3703},[32491],{"type":26,"value":30463},{"type":21,"tag":322,"props":32493,"children":32494},{"style":3638},[32495],{"type":26,"value":3678},{"type":21,"tag":322,"props":32497,"children":32498},{"style":359},[32499],{"type":26,"value":32500},"'matrix_product_state'",{"type":21,"tag":322,"props":32502,"children":32503},{"style":329},[32504],{"type":26,"value":3738},{"type":21,"tag":22,"props":32506,"children":32507},{},[32508,32513],{"type":21,"tag":71,"props":32509,"children":32510},{},[32511],{"type":26,"value":32512},"Aerを使うべき場面:",{"type":26,"value":32514}," あらゆるQiskitのワークフロー、ノイズモデリング、またはIBMハードウェアの挙動を厳密に再現したい場合。",{"type":21,"tag":59,"props":32516,"children":32518},{"id":32517},"pennylane-量子mlに最適",[32519],{"type":26,"value":32520},"PennyLane: 量子MLに最適",{"type":21,"tag":22,"props":32522,"children":32523},{},[32524,32528,32530,32535],{"type":21,"tag":29,"props":32525,"children":32526},{"href":14507},[32527],{"type":26,"value":14223},{"type":26,"value":32529}," は微分可能な回路が必要なときに真価を発揮します。その ",{"type":21,"tag":79,"props":32531,"children":32533},{"className":32532},[],[32534],{"type":26,"value":15855},{"type":26,"value":32536}," シミュレータはパラメータシフト則を用いて厳密な勾配を計算し、量子回路の勾配ベース最適化を可能にします:",{"type":21,"tag":312,"props":32538,"children":32540},{"className":3626,"code":32539,"language":3628,"meta":7,"style":7},"import pennylane as qml\nimport numpy as np\n\ndev = qml.device(\"default.qubit\", wires=2)\n\n@qml.qnode(dev)\ndef circuit(theta):\n    qml.RX(theta[0], wires=0)\n    qml.RY(theta[1], wires=1)\n    qml.CNOT(wires=[0, 1])\n    return qml.expval(qml.PauliZ(0))\n\n# Automatic gradient computation\ngrad_fn = qml.grad(circuit)\ntheta = np.array([0.5, 1.2])\nprint(grad_fn(theta))  # exact gradient\n",[32541],{"type":21,"tag":79,"props":32542,"children":32543},{"__ignoreMap":7},[32544,32563,32582,32589,32628,32635,32646,32662,32695,32727,32762,32781,32788,32796,32813,32847],{"type":21,"tag":322,"props":32545,"children":32546},{"class":324,"line":325},[32547,32551,32555,32559],{"type":21,"tag":322,"props":32548,"children":32549},{"style":3638},[32550],{"type":26,"value":3651},{"type":21,"tag":322,"props":32552,"children":32553},{"style":329},[32554],{"type":26,"value":14530},{"type":21,"tag":322,"props":32556,"children":32557},{"style":3638},[32558],{"type":26,"value":4715},{"type":21,"tag":322,"props":32560,"children":32561},{"style":329},[32562],{"type":26,"value":14539},{"type":21,"tag":322,"props":32564,"children":32565},{"class":324,"line":335},[32566,32570,32574,32578],{"type":21,"tag":322,"props":32567,"children":32568},{"style":3638},[32569],{"type":26,"value":3651},{"type":21,"tag":322,"props":32571,"children":32572},{"style":329},[32573],{"type":26,"value":4710},{"type":21,"tag":322,"props":32575,"children":32576},{"style":3638},[32577],{"type":26,"value":4715},{"type":21,"tag":322,"props":32579,"children":32580},{"style":329},[32581],{"type":26,"value":4720},{"type":21,"tag":322,"props":32583,"children":32584},{"class":324,"line":344},[32585],{"type":21,"tag":322,"props":32586,"children":32587},{"emptyLinePlaceholder":3662},[32588],{"type":26,"value":3665},{"type":21,"tag":322,"props":32590,"children":32591},{"class":324,"line":407},[32592,32596,32600,32604,32608,32612,32616,32620,32624],{"type":21,"tag":322,"props":32593,"children":32594},{"style":329},[32595],{"type":26,"value":14663},{"type":21,"tag":322,"props":32597,"children":32598},{"style":3638},[32599],{"type":26,"value":3678},{"type":21,"tag":322,"props":32601,"children":32602},{"style":329},[32603],{"type":26,"value":14672},{"type":21,"tag":322,"props":32605,"children":32606},{"style":359},[32607],{"type":26,"value":14677},{"type":21,"tag":322,"props":32609,"children":32610},{"style":329},[32611],{"type":26,"value":367},{"type":21,"tag":322,"props":32613,"children":32614},{"style":3703},[32615],{"type":26,"value":14686},{"type":21,"tag":322,"props":32617,"children":32618},{"style":3638},[32619],{"type":26,"value":3678},{"type":21,"tag":322,"props":32621,"children":32622},{"style":348},[32623],{"type":26,"value":423},{"type":21,"tag":322,"props":32625,"children":32626},{"style":329},[32627],{"type":26,"value":3738},{"type":21,"tag":322,"props":32629,"children":32630},{"class":324,"line":431},[32631],{"type":21,"tag":322,"props":32632,"children":32633},{"emptyLinePlaceholder":3662},[32634],{"type":26,"value":3665},{"type":21,"tag":322,"props":32636,"children":32637},{"class":324,"line":453},[32638,32642],{"type":21,"tag":322,"props":32639,"children":32640},{"style":4801},[32641],{"type":26,"value":14710},{"type":21,"tag":322,"props":32643,"children":32644},{"style":329},[32645],{"type":26,"value":14715},{"type":21,"tag":322,"props":32647,"children":32648},{"class":324,"line":474},[32649,32653,32657],{"type":21,"tag":322,"props":32650,"children":32651},{"style":3638},[32652],{"type":26,"value":4798},{"type":21,"tag":322,"props":32654,"children":32655},{"style":4801},[32656],{"type":26,"value":14727},{"type":21,"tag":322,"props":32658,"children":32659},{"style":329},[32660],{"type":26,"value":32661},"(theta):\n",{"type":21,"tag":322,"props":32663,"children":32664},{"class":324,"line":496},[32665,32670,32674,32679,32683,32687,32691],{"type":21,"tag":322,"props":32666,"children":32667},{"style":329},[32668],{"type":26,"value":32669},"    qml.RX(theta[",{"type":21,"tag":322,"props":32671,"children":32672},{"style":348},[32673],{"type":26,"value":5605},{"type":21,"tag":322,"props":32675,"children":32676},{"style":329},[32677],{"type":26,"value":32678},"], ",{"type":21,"tag":322,"props":32680,"children":32681},{"style":3703},[32682],{"type":26,"value":14686},{"type":21,"tag":322,"props":32684,"children":32685},{"style":3638},[32686],{"type":26,"value":3678},{"type":21,"tag":322,"props":32688,"children":32689},{"style":348},[32690],{"type":26,"value":5605},{"type":21,"tag":322,"props":32692,"children":32693},{"style":329},[32694],{"type":26,"value":3738},{"type":21,"tag":322,"props":32696,"children":32697},{"class":324,"line":16},[32698,32703,32707,32711,32715,32719,32723],{"type":21,"tag":322,"props":32699,"children":32700},{"style":329},[32701],{"type":26,"value":32702},"    qml.RY(theta[",{"type":21,"tag":322,"props":32704,"children":32705},{"style":348},[32706],{"type":26,"value":3994},{"type":21,"tag":322,"props":32708,"children":32709},{"style":329},[32710],{"type":26,"value":32678},{"type":21,"tag":322,"props":32712,"children":32713},{"style":3703},[32714],{"type":26,"value":14686},{"type":21,"tag":322,"props":32716,"children":32717},{"style":3638},[32718],{"type":26,"value":3678},{"type":21,"tag":322,"props":32720,"children":32721},{"style":348},[32722],{"type":26,"value":3994},{"type":21,"tag":322,"props":32724,"children":32725},{"style":329},[32726],{"type":26,"value":3738},{"type":21,"tag":322,"props":32728,"children":32729},{"class":324,"line":3884},[32730,32734,32738,32742,32746,32750,32754,32758],{"type":21,"tag":322,"props":32731,"children":32732},{"style":329},[32733],{"type":26,"value":22210},{"type":21,"tag":322,"props":32735,"children":32736},{"style":3703},[32737],{"type":26,"value":14686},{"type":21,"tag":322,"props":32739,"children":32740},{"style":3638},[32741],{"type":26,"value":3678},{"type":21,"tag":322,"props":32743,"children":32744},{"style":329},[32745],{"type":26,"value":22223},{"type":21,"tag":322,"props":32747,"children":32748},{"style":348},[32749],{"type":26,"value":5605},{"type":21,"tag":322,"props":32751,"children":32752},{"style":329},[32753],{"type":26,"value":367},{"type":21,"tag":322,"props":32755,"children":32756},{"style":348},[32757],{"type":26,"value":3994},{"type":21,"tag":322,"props":32759,"children":32760},{"style":329},[32761],{"type":26,"value":3999},{"type":21,"tag":322,"props":32763,"children":32764},{"class":324,"line":3894},[32765,32769,32773,32777],{"type":21,"tag":322,"props":32766,"children":32767},{"style":3638},[32768],{"type":26,"value":5023},{"type":21,"tag":322,"props":32770,"children":32771},{"style":329},[32772],{"type":26,"value":14842},{"type":21,"tag":322,"props":32774,"children":32775},{"style":348},[32776],{"type":26,"value":5605},{"type":21,"tag":322,"props":32778,"children":32779},{"style":329},[32780],{"type":26,"value":12806},{"type":21,"tag":322,"props":32782,"children":32783},{"class":324,"line":1034},[32784],{"type":21,"tag":322,"props":32785,"children":32786},{"emptyLinePlaceholder":3662},[32787],{"type":26,"value":3665},{"type":21,"tag":322,"props":32789,"children":32790},{"class":324,"line":3954},[32791],{"type":21,"tag":322,"props":32792,"children":32793},{"style":3888},[32794],{"type":26,"value":32795},"# Automatic gradient computation\n",{"type":21,"tag":322,"props":32797,"children":32798},{"class":324,"line":4002},[32799,32804,32808],{"type":21,"tag":322,"props":32800,"children":32801},{"style":329},[32802],{"type":26,"value":32803},"grad_fn ",{"type":21,"tag":322,"props":32805,"children":32806},{"style":3638},[32807],{"type":26,"value":3678},{"type":21,"tag":322,"props":32809,"children":32810},{"style":329},[32811],{"type":26,"value":32812}," qml.grad(circuit)\n",{"type":21,"tag":322,"props":32814,"children":32815},{"class":324,"line":4044},[32816,32821,32825,32830,32834,32838,32843],{"type":21,"tag":322,"props":32817,"children":32818},{"style":329},[32819],{"type":26,"value":32820},"theta ",{"type":21,"tag":322,"props":32822,"children":32823},{"style":3638},[32824],{"type":26,"value":3678},{"type":21,"tag":322,"props":32826,"children":32827},{"style":329},[32828],{"type":26,"value":32829}," np.array([",{"type":21,"tag":322,"props":32831,"children":32832},{"style":348},[32833],{"type":26,"value":446},{"type":21,"tag":322,"props":32835,"children":32836},{"style":329},[32837],{"type":26,"value":367},{"type":21,"tag":322,"props":32839,"children":32840},{"style":348},[32841],{"type":26,"value":32842},"1.2",{"type":21,"tag":322,"props":32844,"children":32845},{"style":329},[32846],{"type":26,"value":3999},{"type":21,"tag":322,"props":32848,"children":32849},{"class":324,"line":5068},[32850,32854,32859],{"type":21,"tag":322,"props":32851,"children":32852},{"style":348},[32853],{"type":26,"value":3753},{"type":21,"tag":322,"props":32855,"children":32856},{"style":329},[32857],{"type":26,"value":32858},"(grad_fn(theta))  ",{"type":21,"tag":322,"props":32860,"children":32861},{"style":3888},[32862],{"type":26,"value":32863},"# exact gradient\n",{"type":21,"tag":22,"props":32865,"children":32866},{},[32867,32869,32875,32877,32882],{"type":26,"value":32868},"勾配が不要でより高速なシミュレーションを行うには、",{"type":21,"tag":79,"props":32870,"children":32872},{"className":32871},[],[32873],{"type":26,"value":32874},"lightning.qubit",{"type":26,"value":32876},"（C++バックエンド、",{"type":21,"tag":79,"props":32878,"children":32880},{"className":32879},[],[32881],{"type":26,"value":15855},{"type":26,"value":32883}," より約10倍高速）を使用します:",{"type":21,"tag":312,"props":32885,"children":32887},{"className":10181,"code":32886,"language":10183,"meta":7,"style":7},"pip install pennylane-lightning\n",[32888],{"type":21,"tag":79,"props":32889,"children":32890},{"__ignoreMap":7},[32891],{"type":21,"tag":322,"props":32892,"children":32893},{"class":324,"line":325},[32894,32898,32902],{"type":21,"tag":322,"props":32895,"children":32896},{"style":4801},[32897],{"type":26,"value":12232},{"type":21,"tag":322,"props":32899,"children":32900},{"style":359},[32901],{"type":26,"value":12237},{"type":21,"tag":322,"props":32903,"children":32904},{"style":359},[32905],{"type":26,"value":32906}," pennylane-lightning\n",{"type":21,"tag":312,"props":32908,"children":32910},{"className":3626,"code":32909,"language":3628,"meta":7,"style":7},"dev = qml.device(\"lightning.qubit\", wires=20)\n",[32911],{"type":21,"tag":79,"props":32912,"children":32913},{"__ignoreMap":7},[32914],{"type":21,"tag":322,"props":32915,"children":32916},{"class":324,"line":325},[32917,32921,32925,32929,32934,32938,32942,32946,32951],{"type":21,"tag":322,"props":32918,"children":32919},{"style":329},[32920],{"type":26,"value":14663},{"type":21,"tag":322,"props":32922,"children":32923},{"style":3638},[32924],{"type":26,"value":3678},{"type":21,"tag":322,"props":32926,"children":32927},{"style":329},[32928],{"type":26,"value":14672},{"type":21,"tag":322,"props":32930,"children":32931},{"style":359},[32932],{"type":26,"value":32933},"\"lightning.qubit\"",{"type":21,"tag":322,"props":32935,"children":32936},{"style":329},[32937],{"type":26,"value":367},{"type":21,"tag":322,"props":32939,"children":32940},{"style":3703},[32941],{"type":26,"value":14686},{"type":21,"tag":322,"props":32943,"children":32944},{"style":3638},[32945],{"type":26,"value":3678},{"type":21,"tag":322,"props":32947,"children":32948},{"style":348},[32949],{"type":26,"value":32950},"20",{"type":21,"tag":322,"props":32952,"children":32953},{"style":329},[32954],{"type":26,"value":3738},{"type":21,"tag":22,"props":32956,"children":32957},{},[32958,32963],{"type":21,"tag":71,"props":32959,"children":32960},{},[32961],{"type":26,"value":32962},"PennyLaneを使うべき場面:",{"type":26,"value":32964}," 変分アルゴリズム（VQE、QAOA）、量子ML、回路の勾配を必要とするあらゆるワークフロー。",{"type":21,"tag":59,"props":32966,"children":32968},{"id":32967},"nvidia-cuda-q-速度が重要なとき",[32969],{"type":26,"value":32970},"NVIDIA CUDA-Q: 速度が重要なとき",{"type":21,"tag":22,"props":32972,"children":32973},{},[32974,32978],{"type":21,"tag":29,"props":32975,"children":32976},{"href":21050},[32977],{"type":26,"value":18654},{"type":26,"value":32979}," はGPUによって高速化されたシミュレーションを提供し、大規模回路ではCPUシミュレータの100〜10,000倍高速になり得ます。NVIDIA GPUをお持ちなら、25量子ビット以上の回路にはこれが正しい選択です:",{"type":21,"tag":312,"props":32981,"children":32983},{"className":3626,"code":32982,"language":3628,"meta":7,"style":7},"import cudaq\n\n@cudaq.kernel\ndef large_circuit(n: int):\n    qvec = cudaq.qvector(n)\n    h(qvec[0])\n    for i in range(n - 1):\n        cx(qvec[i], qvec[i + 1])\n    mz(qvec)\n\n# Run on GPU (specify 'nvidia' target)\ncudaq.set_target('nvidia')\ncounts = cudaq.sample(large_circuit, 30, shots_count=10000)\nprint(counts)\n",[32984],{"type":21,"tag":79,"props":32985,"children":32986},{"__ignoreMap":7},[32987,32998,33005,33012,33037,33053,33068,33103,33123,33131,33138,33146,33163,33205],{"type":21,"tag":322,"props":32988,"children":32989},{"class":324,"line":325},[32990,32994],{"type":21,"tag":322,"props":32991,"children":32992},{"style":3638},[32993],{"type":26,"value":3651},{"type":21,"tag":322,"props":32995,"children":32996},{"style":329},[32997],{"type":26,"value":22323},{"type":21,"tag":322,"props":32999,"children":33000},{"class":324,"line":335},[33001],{"type":21,"tag":322,"props":33002,"children":33003},{"emptyLinePlaceholder":3662},[33004],{"type":26,"value":3665},{"type":21,"tag":322,"props":33006,"children":33007},{"class":324,"line":344},[33008],{"type":21,"tag":322,"props":33009,"children":33010},{"style":4801},[33011],{"type":26,"value":22331},{"type":21,"tag":322,"props":33013,"children":33014},{"class":324,"line":407},[33015,33019,33024,33029,33033],{"type":21,"tag":322,"props":33016,"children":33017},{"style":3638},[33018],{"type":26,"value":4798},{"type":21,"tag":322,"props":33020,"children":33021},{"style":4801},[33022],{"type":26,"value":33023}," large_circuit",{"type":21,"tag":322,"props":33025,"children":33026},{"style":329},[33027],{"type":26,"value":33028},"(n: ",{"type":21,"tag":322,"props":33030,"children":33031},{"style":348},[33032],{"type":26,"value":5460},{"type":21,"tag":322,"props":33034,"children":33035},{"style":329},[33036],{"type":26,"value":4851},{"type":21,"tag":322,"props":33038,"children":33039},{"class":324,"line":431},[33040,33044,33048],{"type":21,"tag":322,"props":33041,"children":33042},{"style":329},[33043],{"type":26,"value":22354},{"type":21,"tag":322,"props":33045,"children":33046},{"style":3638},[33047],{"type":26,"value":3678},{"type":21,"tag":322,"props":33049,"children":33050},{"style":329},[33051],{"type":26,"value":33052}," cudaq.qvector(n)\n",{"type":21,"tag":322,"props":33054,"children":33055},{"class":324,"line":453},[33056,33060,33064],{"type":21,"tag":322,"props":33057,"children":33058},{"style":329},[33059],{"type":26,"value":22379},{"type":21,"tag":322,"props":33061,"children":33062},{"style":348},[33063],{"type":26,"value":5605},{"type":21,"tag":322,"props":33065,"children":33066},{"style":329},[33067],{"type":26,"value":3999},{"type":21,"tag":322,"props":33069,"children":33070},{"class":324,"line":474},[33071,33075,33079,33083,33087,33091,33095,33099],{"type":21,"tag":322,"props":33072,"children":33073},{"style":3638},[33074],{"type":26,"value":5177},{"type":21,"tag":322,"props":33076,"children":33077},{"style":329},[33078],{"type":26,"value":5182},{"type":21,"tag":322,"props":33080,"children":33081},{"style":3638},[33082],{"type":26,"value":3941},{"type":21,"tag":322,"props":33084,"children":33085},{"style":348},[33086],{"type":26,"value":3946},{"type":21,"tag":322,"props":33088,"children":33089},{"style":329},[33090],{"type":26,"value":8106},{"type":21,"tag":322,"props":33092,"children":33093},{"style":3638},[33094],{"type":26,"value":5285},{"type":21,"tag":322,"props":33096,"children":33097},{"style":348},[33098],{"type":26,"value":6536},{"type":21,"tag":322,"props":33100,"children":33101},{"style":329},[33102],{"type":26,"value":4851},{"type":21,"tag":322,"props":33104,"children":33105},{"class":324,"line":496},[33106,33111,33115,33119],{"type":21,"tag":322,"props":33107,"children":33108},{"style":329},[33109],{"type":26,"value":33110},"        cx(qvec[i], qvec[i ",{"type":21,"tag":322,"props":33112,"children":33113},{"style":3638},[33114],{"type":26,"value":6965},{"type":21,"tag":322,"props":33116,"children":33117},{"style":348},[33118],{"type":26,"value":6536},{"type":21,"tag":322,"props":33120,"children":33121},{"style":329},[33122],{"type":26,"value":3999},{"type":21,"tag":322,"props":33124,"children":33125},{"class":324,"line":16},[33126],{"type":21,"tag":322,"props":33127,"children":33128},{"style":329},[33129],{"type":26,"value":33130},"    mz(qvec)\n",{"type":21,"tag":322,"props":33132,"children":33133},{"class":324,"line":3884},[33134],{"type":21,"tag":322,"props":33135,"children":33136},{"emptyLinePlaceholder":3662},[33137],{"type":26,"value":3665},{"type":21,"tag":322,"props":33139,"children":33140},{"class":324,"line":3894},[33141],{"type":21,"tag":322,"props":33142,"children":33143},{"style":3888},[33144],{"type":26,"value":33145},"# Run on GPU (specify 'nvidia' target)\n",{"type":21,"tag":322,"props":33147,"children":33148},{"class":324,"line":1034},[33149,33154,33159],{"type":21,"tag":322,"props":33150,"children":33151},{"style":329},[33152],{"type":26,"value":33153},"cudaq.set_target(",{"type":21,"tag":322,"props":33155,"children":33156},{"style":359},[33157],{"type":26,"value":33158},"'nvidia'",{"type":21,"tag":322,"props":33160,"children":33161},{"style":329},[33162],{"type":26,"value":3738},{"type":21,"tag":322,"props":33164,"children":33165},{"class":324,"line":3954},[33166,33170,33174,33179,33184,33188,33192,33196,33201],{"type":21,"tag":322,"props":33167,"children":33168},{"style":329},[33169],{"type":26,"value":8894},{"type":21,"tag":322,"props":33171,"children":33172},{"style":3638},[33173],{"type":26,"value":3678},{"type":21,"tag":322,"props":33175,"children":33176},{"style":329},[33177],{"type":26,"value":33178}," cudaq.sample(large_circuit, ",{"type":21,"tag":322,"props":33180,"children":33181},{"style":348},[33182],{"type":26,"value":33183},"30",{"type":21,"tag":322,"props":33185,"children":33186},{"style":329},[33187],{"type":26,"value":367},{"type":21,"tag":322,"props":33189,"children":33190},{"style":3703},[33191],{"type":26,"value":22428},{"type":21,"tag":322,"props":33193,"children":33194},{"style":3638},[33195],{"type":26,"value":3678},{"type":21,"tag":322,"props":33197,"children":33198},{"style":348},[33199],{"type":26,"value":33200},"10000",{"type":21,"tag":322,"props":33202,"children":33203},{"style":329},[33204],{"type":26,"value":3738},{"type":21,"tag":322,"props":33206,"children":33207},{"class":324,"line":4002},[33208,33212],{"type":21,"tag":322,"props":33209,"children":33210},{"style":348},[33211],{"type":26,"value":3753},{"type":21,"tag":322,"props":33213,"children":33214},{"style":329},[33215],{"type":26,"value":33216},"(counts)\n",{"type":21,"tag":22,"props":33218,"children":33219},{},[33220,33225],{"type":21,"tag":71,"props":33221,"children":33222},{},[33223],{"type":26,"value":33224},"性能ベンチマーク",{"type":26,"value":33226},"（GHZ回路、28量子ビット、1000ショット）:",{"type":21,"tag":3556,"props":33228,"children":33229},{},[33230,33235,33240],{"type":21,"tag":3560,"props":33231,"children":33232},{},[33233],{"type":26,"value":33234},"Qiskit Aer CPU: 約45秒",{"type":21,"tag":3560,"props":33236,"children":33237},{},[33238],{"type":26,"value":33239},"PennyLane lightning.qubit: 約30秒",{"type":21,"tag":3560,"props":33241,"children":33242},{},[33243],{"type":26,"value":33244},"CUDA-Q（A100 GPU）: 約0.8秒",{"type":21,"tag":22,"props":33246,"children":33247},{},[33248,33253],{"type":21,"tag":71,"props":33249,"children":33250},{},[33251],{"type":26,"value":33252},"CUDA-Qを使うべき場面:",{"type":26,"value":33254}," 25量子ビット以上のあらゆる回路、性能が重要なシミュレーション、マルチGPUワークロード。",{"type":21,"tag":59,"props":33256,"children":33258},{"id":33257},"cirq-ノイズモデリングとnisq研究",[33259],{"type":26,"value":33260},"Cirq: ノイズモデリングとNISQ研究",{"type":21,"tag":22,"props":33262,"children":33263},{},[33264,33268],{"type":21,"tag":29,"props":33265,"children":33266},{"href":9270},[33267],{"type":26,"value":21280},{"type":26,"value":33269}," には3つのシミュレータが含まれています:",{"type":21,"tag":312,"props":33271,"children":33273},{"className":3626,"code":33272,"language":3628,"meta":7,"style":7},"import cirq\n\n# Exact statevector simulation\nsim = cirq.Simulator()\n\n# Density matrix with noise\nnoise_model = cirq.ConstantQubitNoiseModel(\n    cirq.depolarize(p=0.01)\n)\nnoisy_sim = cirq.DensityMatrixSimulator(noise=noise_model)\n\n# Clifford circuits only (but exponentially faster for stabilizer states)\nclifford_sim = cirq.CliffordSimulator()\n",[33274],{"type":21,"tag":79,"props":33275,"children":33276},{"__ignoreMap":7},[33277,33289,33296,33304,33320,33327,33335,33352,33377,33384,33414,33421,33429],{"type":21,"tag":322,"props":33278,"children":33279},{"class":324,"line":325},[33280,33284],{"type":21,"tag":322,"props":33281,"children":33282},{"style":3638},[33283],{"type":26,"value":3651},{"type":21,"tag":322,"props":33285,"children":33286},{"style":329},[33287],{"type":26,"value":33288}," cirq\n",{"type":21,"tag":322,"props":33290,"children":33291},{"class":324,"line":335},[33292],{"type":21,"tag":322,"props":33293,"children":33294},{"emptyLinePlaceholder":3662},[33295],{"type":26,"value":3665},{"type":21,"tag":322,"props":33297,"children":33298},{"class":324,"line":344},[33299],{"type":21,"tag":322,"props":33300,"children":33301},{"style":3888},[33302],{"type":26,"value":33303},"# Exact statevector simulation\n",{"type":21,"tag":322,"props":33305,"children":33306},{"class":324,"line":407},[33307,33311,33315],{"type":21,"tag":322,"props":33308,"children":33309},{"style":329},[33310],{"type":26,"value":8861},{"type":21,"tag":322,"props":33312,"children":33313},{"style":3638},[33314],{"type":26,"value":3678},{"type":21,"tag":322,"props":33316,"children":33317},{"style":329},[33318],{"type":26,"value":33319}," cirq.Simulator()\n",{"type":21,"tag":322,"props":33321,"children":33322},{"class":324,"line":431},[33323],{"type":21,"tag":322,"props":33324,"children":33325},{"emptyLinePlaceholder":3662},[33326],{"type":26,"value":3665},{"type":21,"tag":322,"props":33328,"children":33329},{"class":324,"line":453},[33330],{"type":21,"tag":322,"props":33331,"children":33332},{"style":3888},[33333],{"type":26,"value":33334},"# Density matrix with noise\n",{"type":21,"tag":322,"props":33336,"children":33337},{"class":324,"line":474},[33338,33343,33347],{"type":21,"tag":322,"props":33339,"children":33340},{"style":329},[33341],{"type":26,"value":33342},"noise_model ",{"type":21,"tag":322,"props":33344,"children":33345},{"style":3638},[33346],{"type":26,"value":3678},{"type":21,"tag":322,"props":33348,"children":33349},{"style":329},[33350],{"type":26,"value":33351}," cirq.ConstantQubitNoiseModel(\n",{"type":21,"tag":322,"props":33353,"children":33354},{"class":324,"line":496},[33355,33360,33364,33368,33373],{"type":21,"tag":322,"props":33356,"children":33357},{"style":329},[33358],{"type":26,"value":33359},"    cirq.depolarize(",{"type":21,"tag":322,"props":33361,"children":33362},{"style":3703},[33363],{"type":26,"value":22},{"type":21,"tag":322,"props":33365,"children":33366},{"style":3638},[33367],{"type":26,"value":3678},{"type":21,"tag":322,"props":33369,"children":33370},{"style":348},[33371],{"type":26,"value":33372},"0.01",{"type":21,"tag":322,"props":33374,"children":33375},{"style":329},[33376],{"type":26,"value":3738},{"type":21,"tag":322,"props":33378,"children":33379},{"class":324,"line":16},[33380],{"type":21,"tag":322,"props":33381,"children":33382},{"style":329},[33383],{"type":26,"value":3738},{"type":21,"tag":322,"props":33385,"children":33386},{"class":324,"line":3884},[33387,33392,33396,33401,33405,33409],{"type":21,"tag":322,"props":33388,"children":33389},{"style":329},[33390],{"type":26,"value":33391},"noisy_sim ",{"type":21,"tag":322,"props":33393,"children":33394},{"style":3638},[33395],{"type":26,"value":3678},{"type":21,"tag":322,"props":33397,"children":33398},{"style":329},[33399],{"type":26,"value":33400}," cirq.DensityMatrixSimulator(",{"type":21,"tag":322,"props":33402,"children":33403},{"style":3703},[33404],{"type":26,"value":15077},{"type":21,"tag":322,"props":33406,"children":33407},{"style":3638},[33408],{"type":26,"value":3678},{"type":21,"tag":322,"props":33410,"children":33411},{"style":329},[33412],{"type":26,"value":33413},"noise_model)\n",{"type":21,"tag":322,"props":33415,"children":33416},{"class":324,"line":3894},[33417],{"type":21,"tag":322,"props":33418,"children":33419},{"emptyLinePlaceholder":3662},[33420],{"type":26,"value":3665},{"type":21,"tag":322,"props":33422,"children":33423},{"class":324,"line":1034},[33424],{"type":21,"tag":322,"props":33425,"children":33426},{"style":3888},[33427],{"type":26,"value":33428},"# Clifford circuits only (but exponentially faster for stabilizer states)\n",{"type":21,"tag":322,"props":33430,"children":33431},{"class":324,"line":3954},[33432,33437,33441],{"type":21,"tag":322,"props":33433,"children":33434},{"style":329},[33435],{"type":26,"value":33436},"clifford_sim ",{"type":21,"tag":322,"props":33438,"children":33439},{"style":3638},[33440],{"type":26,"value":3678},{"type":21,"tag":322,"props":33442,"children":33443},{"style":329},[33444],{"type":26,"value":33445}," cirq.CliffordSimulator()\n",{"type":21,"tag":22,"props":33447,"children":33448},{},[33449,33454],{"type":21,"tag":71,"props":33450,"children":33451},{},[33452],{"type":26,"value":33453},"Cirqを使うべき場面:",{"type":26,"value":33455}," NISQのノイズモデリング、スタビライザ回路の研究、Google AI Quantumのワークフロー。",{"type":21,"tag":59,"props":33457,"children":33459},{"id":33458},"amazon-braket-localsimulator-隔離性と可搬性",[33460],{"type":26,"value":33461},"Amazon Braket LocalSimulator: 隔離性と可搬性",{"type":21,"tag":22,"props":33463,"children":33464},{},[33465,33471],{"type":21,"tag":29,"props":33466,"children":33468},{"href":33467},"\u002Fsdks\u002Fbraket",[33469],{"type":26,"value":33470},"Braket SDK",{"type":26,"value":33472}," には、クラウドAPIを厳密に再現する無料のローカルシミュレータが含まれています:",{"type":21,"tag":312,"props":33474,"children":33476},{"className":3626,"code":33475,"language":3628,"meta":7,"style":7},"from braket.devices import LocalSimulator\nfrom braket.circuits import Circuit\n\ndevice = LocalSimulator()\n\ncircuit = Circuit()\ncircuit.h(0)\ncircuit.cnot(0, 1)\ncircuit.probability()\n\ntask = device.run(circuit, shots=1000)\nresult = task.result()\nprint(result.measurement_counts)\n",[33477],{"type":21,"tag":79,"props":33478,"children":33479},{"__ignoreMap":7},[33480,33501,33522,33529,33546,33553,33570,33586,33610,33618,33625,33658,33674],{"type":21,"tag":322,"props":33481,"children":33482},{"class":324,"line":325},[33483,33487,33492,33496],{"type":21,"tag":322,"props":33484,"children":33485},{"style":3638},[33486],{"type":26,"value":3641},{"type":21,"tag":322,"props":33488,"children":33489},{"style":329},[33490],{"type":26,"value":33491}," braket.devices ",{"type":21,"tag":322,"props":33493,"children":33494},{"style":3638},[33495],{"type":26,"value":3651},{"type":21,"tag":322,"props":33497,"children":33498},{"style":329},[33499],{"type":26,"value":33500}," LocalSimulator\n",{"type":21,"tag":322,"props":33502,"children":33503},{"class":324,"line":335},[33504,33508,33513,33517],{"type":21,"tag":322,"props":33505,"children":33506},{"style":3638},[33507],{"type":26,"value":3641},{"type":21,"tag":322,"props":33509,"children":33510},{"style":329},[33511],{"type":26,"value":33512}," braket.circuits ",{"type":21,"tag":322,"props":33514,"children":33515},{"style":3638},[33516],{"type":26,"value":3651},{"type":21,"tag":322,"props":33518,"children":33519},{"style":329},[33520],{"type":26,"value":33521}," Circuit\n",{"type":21,"tag":322,"props":33523,"children":33524},{"class":324,"line":344},[33525],{"type":21,"tag":322,"props":33526,"children":33527},{"emptyLinePlaceholder":3662},[33528],{"type":26,"value":3665},{"type":21,"tag":322,"props":33530,"children":33531},{"class":324,"line":407},[33532,33537,33541],{"type":21,"tag":322,"props":33533,"children":33534},{"style":329},[33535],{"type":26,"value":33536},"device ",{"type":21,"tag":322,"props":33538,"children":33539},{"style":3638},[33540],{"type":26,"value":3678},{"type":21,"tag":322,"props":33542,"children":33543},{"style":329},[33544],{"type":26,"value":33545}," LocalSimulator()\n",{"type":21,"tag":322,"props":33547,"children":33548},{"class":324,"line":431},[33549],{"type":21,"tag":322,"props":33550,"children":33551},{"emptyLinePlaceholder":3662},[33552],{"type":26,"value":3665},{"type":21,"tag":322,"props":33554,"children":33555},{"class":324,"line":453},[33556,33561,33565],{"type":21,"tag":322,"props":33557,"children":33558},{"style":329},[33559],{"type":26,"value":33560},"circuit ",{"type":21,"tag":322,"props":33562,"children":33563},{"style":3638},[33564],{"type":26,"value":3678},{"type":21,"tag":322,"props":33566,"children":33567},{"style":329},[33568],{"type":26,"value":33569}," Circuit()\n",{"type":21,"tag":322,"props":33571,"children":33572},{"class":324,"line":474},[33573,33578,33582],{"type":21,"tag":322,"props":33574,"children":33575},{"style":329},[33576],{"type":26,"value":33577},"circuit.h(",{"type":21,"tag":322,"props":33579,"children":33580},{"style":348},[33581],{"type":26,"value":5605},{"type":21,"tag":322,"props":33583,"children":33584},{"style":329},[33585],{"type":26,"value":3738},{"type":21,"tag":322,"props":33587,"children":33588},{"class":324,"line":496},[33589,33594,33598,33602,33606],{"type":21,"tag":322,"props":33590,"children":33591},{"style":329},[33592],{"type":26,"value":33593},"circuit.cnot(",{"type":21,"tag":322,"props":33595,"children":33596},{"style":348},[33597],{"type":26,"value":5605},{"type":21,"tag":322,"props":33599,"children":33600},{"style":329},[33601],{"type":26,"value":367},{"type":21,"tag":322,"props":33603,"children":33604},{"style":348},[33605],{"type":26,"value":3994},{"type":21,"tag":322,"props":33607,"children":33608},{"style":329},[33609],{"type":26,"value":3738},{"type":21,"tag":322,"props":33611,"children":33612},{"class":324,"line":16},[33613],{"type":21,"tag":322,"props":33614,"children":33615},{"style":329},[33616],{"type":26,"value":33617},"circuit.probability()\n",{"type":21,"tag":322,"props":33619,"children":33620},{"class":324,"line":3884},[33621],{"type":21,"tag":322,"props":33622,"children":33623},{"emptyLinePlaceholder":3662},[33624],{"type":26,"value":3665},{"type":21,"tag":322,"props":33626,"children":33627},{"class":324,"line":3894},[33628,33633,33637,33642,33646,33650,33654],{"type":21,"tag":322,"props":33629,"children":33630},{"style":329},[33631],{"type":26,"value":33632},"task ",{"type":21,"tag":322,"props":33634,"children":33635},{"style":3638},[33636],{"type":26,"value":3678},{"type":21,"tag":322,"props":33638,"children":33639},{"style":329},[33640],{"type":26,"value":33641}," device.run(circuit, ",{"type":21,"tag":322,"props":33643,"children":33644},{"style":3703},[33645],{"type":26,"value":4560},{"type":21,"tag":322,"props":33647,"children":33648},{"style":3638},[33649],{"type":26,"value":3678},{"type":21,"tag":322,"props":33651,"children":33652},{"style":348},[33653],{"type":26,"value":20897},{"type":21,"tag":322,"props":33655,"children":33656},{"style":329},[33657],{"type":26,"value":3738},{"type":21,"tag":322,"props":33659,"children":33660},{"class":324,"line":1034},[33661,33665,33669],{"type":21,"tag":322,"props":33662,"children":33663},{"style":329},[33664],{"type":26,"value":19764},{"type":21,"tag":322,"props":33666,"children":33667},{"style":3638},[33668],{"type":26,"value":3678},{"type":21,"tag":322,"props":33670,"children":33671},{"style":329},[33672],{"type":26,"value":33673}," task.result()\n",{"type":21,"tag":322,"props":33675,"children":33676},{"class":324,"line":3954},[33677,33681],{"type":21,"tag":322,"props":33678,"children":33679},{"style":348},[33680],{"type":26,"value":3753},{"type":21,"tag":322,"props":33682,"children":33683},{"style":329},[33684],{"type":26,"value":33685},"(result.measurement_counts)\n",{"type":21,"tag":22,"props":33687,"children":33688},{},[33689,33694],{"type":21,"tag":71,"props":33690,"children":33691},{},[33692],{"type":26,"value":33693},"Braketローカルを使うべき場面:",{"type":26,"value":33695}," AWSへのデプロイを前提に開発しており、クラウドAPIと一致するローカルの開発\u002Fテスト環境が欲しい場合。",{"type":21,"tag":59,"props":33697,"children":33699},{"id":33698},"hlquantumという近道",[33700],{"type":26,"value":33701},"HLQuantumという近道",{"type":21,"tag":22,"props":33703,"children":33704},{},[33705,33707,33711],{"type":26,"value":33706},"同じ回路を複数のシミュレータで実行してベンチマークや比較を行う必要がある場合、",{"type":21,"tag":29,"props":33708,"children":33709},{"href":19441},[33710],{"type":26,"value":19444},{"type":26,"value":33712}," がそれを簡単にしてくれます:",{"type":21,"tag":312,"props":33714,"children":33716},{"className":3626,"code":33715,"language":3628,"meta":7,"style":7},"import hlquantum as hlq\nimport time\n\nqc = hlq.Circuit(28)\nqc.h(0)\nfor i in range(27):\n    qc.cx(i, i + 1)\nqc.measure_all()\n\nfor backend in [\"qiskit\", \"pennylane\", \"cudaq\"]:\n    t0 = time.time()\n    result = hlq.run(qc, shots=1000, backend=backend)\n    print(f\"{backend}: {time.time() - t0:.2f}s\")\n",[33717],{"type":21,"tag":79,"props":33718,"children":33719},{"__ignoreMap":7},[33720,33739,33751,33758,33782,33797,33829,33849,33856,33863,33909,33926,33970],{"type":21,"tag":322,"props":33721,"children":33722},{"class":324,"line":325},[33723,33727,33731,33735],{"type":21,"tag":322,"props":33724,"children":33725},{"style":3638},[33726],{"type":26,"value":3651},{"type":21,"tag":322,"props":33728,"children":33729},{"style":329},[33730],{"type":26,"value":19732},{"type":21,"tag":322,"props":33732,"children":33733},{"style":3638},[33734],{"type":26,"value":4715},{"type":21,"tag":322,"props":33736,"children":33737},{"style":329},[33738],{"type":26,"value":19741},{"type":21,"tag":322,"props":33740,"children":33741},{"class":324,"line":335},[33742,33746],{"type":21,"tag":322,"props":33743,"children":33744},{"style":3638},[33745],{"type":26,"value":3651},{"type":21,"tag":322,"props":33747,"children":33748},{"style":329},[33749],{"type":26,"value":33750}," time\n",{"type":21,"tag":322,"props":33752,"children":33753},{"class":324,"line":344},[33754],{"type":21,"tag":322,"props":33755,"children":33756},{"emptyLinePlaceholder":3662},[33757],{"type":26,"value":3665},{"type":21,"tag":322,"props":33759,"children":33760},{"class":324,"line":407},[33761,33765,33769,33773,33778],{"type":21,"tag":322,"props":33762,"children":33763},{"style":329},[33764],{"type":26,"value":7819},{"type":21,"tag":322,"props":33766,"children":33767},{"style":3638},[33768],{"type":26,"value":3678},{"type":21,"tag":322,"props":33770,"children":33771},{"style":329},[33772],{"type":26,"value":21793},{"type":21,"tag":322,"props":33774,"children":33775},{"style":348},[33776],{"type":26,"value":33777},"28",{"type":21,"tag":322,"props":33779,"children":33780},{"style":329},[33781],{"type":26,"value":3738},{"type":21,"tag":322,"props":33783,"children":33784},{"class":324,"line":431},[33785,33789,33793],{"type":21,"tag":322,"props":33786,"children":33787},{"style":329},[33788],{"type":26,"value":7835},{"type":21,"tag":322,"props":33790,"children":33791},{"style":348},[33792],{"type":26,"value":5605},{"type":21,"tag":322,"props":33794,"children":33795},{"style":329},[33796],{"type":26,"value":3738},{"type":21,"tag":322,"props":33798,"children":33799},{"class":324,"line":453},[33800,33804,33808,33812,33816,33820,33825],{"type":21,"tag":322,"props":33801,"children":33802},{"style":3638},[33803],{"type":26,"value":3931},{"type":21,"tag":322,"props":33805,"children":33806},{"style":329},[33807],{"type":26,"value":5182},{"type":21,"tag":322,"props":33809,"children":33810},{"style":3638},[33811],{"type":26,"value":3941},{"type":21,"tag":322,"props":33813,"children":33814},{"style":348},[33815],{"type":26,"value":3946},{"type":21,"tag":322,"props":33817,"children":33818},{"style":329},[33819],{"type":26,"value":3758},{"type":21,"tag":322,"props":33821,"children":33822},{"style":348},[33823],{"type":26,"value":33824},"27",{"type":21,"tag":322,"props":33826,"children":33827},{"style":329},[33828],{"type":26,"value":4851},{"type":21,"tag":322,"props":33830,"children":33831},{"class":324,"line":474},[33832,33837,33841,33845],{"type":21,"tag":322,"props":33833,"children":33834},{"style":329},[33835],{"type":26,"value":33836},"    qc.cx(i, i ",{"type":21,"tag":322,"props":33838,"children":33839},{"style":3638},[33840],{"type":26,"value":6965},{"type":21,"tag":322,"props":33842,"children":33843},{"style":348},[33844],{"type":26,"value":6536},{"type":21,"tag":322,"props":33846,"children":33847},{"style":329},[33848],{"type":26,"value":3738},{"type":21,"tag":322,"props":33850,"children":33851},{"class":324,"line":496},[33852],{"type":21,"tag":322,"props":33853,"children":33854},{"style":329},[33855],{"type":26,"value":13527},{"type":21,"tag":322,"props":33857,"children":33858},{"class":324,"line":16},[33859],{"type":21,"tag":322,"props":33860,"children":33861},{"emptyLinePlaceholder":3662},[33862],{"type":26,"value":3665},{"type":21,"tag":322,"props":33864,"children":33865},{"class":324,"line":3884},[33866,33870,33875,33879,33883,33887,33891,33896,33900,33905],{"type":21,"tag":322,"props":33867,"children":33868},{"style":3638},[33869],{"type":26,"value":3931},{"type":21,"tag":322,"props":33871,"children":33872},{"style":329},[33873],{"type":26,"value":33874}," backend ",{"type":21,"tag":322,"props":33876,"children":33877},{"style":3638},[33878],{"type":26,"value":3941},{"type":21,"tag":322,"props":33880,"children":33881},{"style":329},[33882],{"type":26,"value":7098},{"type":21,"tag":322,"props":33884,"children":33885},{"style":359},[33886],{"type":26,"value":29206},{"type":21,"tag":322,"props":33888,"children":33889},{"style":329},[33890],{"type":26,"value":367},{"type":21,"tag":322,"props":33892,"children":33893},{"style":359},[33894],{"type":26,"value":33895},"\"pennylane\"",{"type":21,"tag":322,"props":33897,"children":33898},{"style":329},[33899],{"type":26,"value":367},{"type":21,"tag":322,"props":33901,"children":33902},{"style":359},[33903],{"type":26,"value":33904},"\"cudaq\"",{"type":21,"tag":322,"props":33906,"children":33907},{"style":329},[33908],{"type":26,"value":18024},{"type":21,"tag":322,"props":33910,"children":33911},{"class":324,"line":3894},[33912,33917,33921],{"type":21,"tag":322,"props":33913,"children":33914},{"style":329},[33915],{"type":26,"value":33916},"    t0 ",{"type":21,"tag":322,"props":33918,"children":33919},{"style":3638},[33920],{"type":26,"value":3678},{"type":21,"tag":322,"props":33922,"children":33923},{"style":329},[33924],{"type":26,"value":33925}," time.time()\n",{"type":21,"tag":322,"props":33927,"children":33928},{"class":324,"line":1034},[33929,33934,33938,33942,33946,33950,33954,33958,33962,33966],{"type":21,"tag":322,"props":33930,"children":33931},{"style":329},[33932],{"type":26,"value":33933},"    result ",{"type":21,"tag":322,"props":33935,"children":33936},{"style":3638},[33937],{"type":26,"value":3678},{"type":21,"tag":322,"props":33939,"children":33940},{"style":329},[33941],{"type":26,"value":21857},{"type":21,"tag":322,"props":33943,"children":33944},{"style":3703},[33945],{"type":26,"value":4560},{"type":21,"tag":322,"props":33947,"children":33948},{"style":3638},[33949],{"type":26,"value":3678},{"type":21,"tag":322,"props":33951,"children":33952},{"style":348},[33953],{"type":26,"value":20897},{"type":21,"tag":322,"props":33955,"children":33956},{"style":329},[33957],{"type":26,"value":367},{"type":21,"tag":322,"props":33959,"children":33960},{"style":3703},[33961],{"type":26,"value":18908},{"type":21,"tag":322,"props":33963,"children":33964},{"style":3638},[33965],{"type":26,"value":3678},{"type":21,"tag":322,"props":33967,"children":33968},{"style":329},[33969],{"type":26,"value":28725},{"type":21,"tag":322,"props":33971,"children":33972},{"class":324,"line":3954},[33973,33977,33981,33985,33989,33993,33997,34001,34005,34009,34014,34018,34023,34028,34032,34037],{"type":21,"tag":322,"props":33974,"children":33975},{"style":348},[33976],{"type":26,"value":17663},{"type":21,"tag":322,"props":33978,"children":33979},{"style":329},[33980],{"type":26,"value":3758},{"type":21,"tag":322,"props":33982,"children":33983},{"style":3638},[33984],{"type":26,"value":3763},{"type":21,"tag":322,"props":33986,"children":33987},{"style":359},[33988],{"type":26,"value":3788},{"type":21,"tag":322,"props":33990,"children":33991},{"style":348},[33992],{"type":26,"value":3773},{"type":21,"tag":322,"props":33994,"children":33995},{"style":329},[33996],{"type":26,"value":18908},{"type":21,"tag":322,"props":33998,"children":33999},{"style":348},[34000],{"type":26,"value":3783},{"type":21,"tag":322,"props":34002,"children":34003},{"style":359},[34004],{"type":26,"value":418},{"type":21,"tag":322,"props":34006,"children":34007},{"style":348},[34008],{"type":26,"value":3773},{"type":21,"tag":322,"props":34010,"children":34011},{"style":329},[34012],{"type":26,"value":34013},"time.time() ",{"type":21,"tag":322,"props":34015,"children":34016},{"style":3638},[34017],{"type":26,"value":5285},{"type":21,"tag":322,"props":34019,"children":34020},{"style":329},[34021],{"type":26,"value":34022}," t0",{"type":21,"tag":322,"props":34024,"children":34025},{"style":3638},[34026],{"type":26,"value":34027},":.2f",{"type":21,"tag":322,"props":34029,"children":34030},{"style":348},[34031],{"type":26,"value":3783},{"type":21,"tag":322,"props":34033,"children":34034},{"style":359},[34035],{"type":26,"value":34036},"s\"",{"type":21,"tag":322,"props":34038,"children":34039},{"style":329},[34040],{"type":26,"value":3738},{"type":21,"tag":22,"props":34042,"children":34043},{},[34044],{"type":26,"value":34045},"1つの回路、3つのバックエンド、そのまま比較できる結果。",{"type":21,"tag":703,"props":34047,"children":34048},{},[34049],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":34051},[34052,34053,34054,34055,34056,34057,34058],{"id":32081,"depth":335,"text":32081},{"id":32269,"depth":335,"text":32272},{"id":32517,"depth":335,"text":32520},{"id":32967,"depth":335,"text":32970},{"id":33257,"depth":335,"text":33260},{"id":33458,"depth":335,"text":33461},{"id":33698,"depth":335,"text":33701},"content:ja:blog:quantum-simulators-compared.md","ja\u002Fblog\u002Fquantum-simulators-compared.md","ja\u002Fblog\u002Fquantum-simulators-compared",{"_path":34063,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":34064,"description":34065,"date":34066,"author":11,"tags":34067,"readingTime":16,"body":34068,"_type":715,"_id":36161,"_source":717,"_file":36162,"_stem":36163,"_extension":720},"\u002Fja\u002Fblog\u002Fvqe-pennylane-practical-guide","PennyLane で VQE を構築する: 実践ガイド","PennyLane を用いた変分量子固有値ソルバーのステップバイステップ実装 — ハミルトニアンの設定から古典的最適化ループまで。","2026-01-28",[883,14223,7585,14222],{"type":18,"children":34069,"toc":36152},[34070,34075,34081,34086,34109,34114,34119,34124,34151,34156,34487,34492,34502,34832,34838,34843,35318,35323,35331,35336,35341,35346,35630,35636,35641,35989,35994,35999,36009,36019,36037,36130,36148],{"type":21,"tag":22,"props":34071,"children":34072},{},[34073],{"type":26,"value":34074},"変分量子固有値ソルバー (VQE) は、最も重要な近未来の量子アルゴリズムの一つです。これは分子や材料の基底状態エネルギーを求めます — この計算は古典コンピュータにとっては指数的に困難ですが、NISQ デバイス上では扱うことができます。本ガイドでは、PennyLane による完全な VQE 実装を順を追って解説します。",{"type":21,"tag":59,"props":34076,"children":34078},{"id":34077},"vqe-が行うこと",[34079],{"type":26,"value":34080},"VQE が行うこと",{"type":21,"tag":22,"props":34082,"children":34083},{},[34084],{"type":26,"value":34085},"VQE はハミルトニアン H (通常は分子のエネルギーを表す) の最小固有値を求めます。その仕組みは次の通りです:",{"type":21,"tag":7444,"props":34087,"children":34088},{},[34089,34094,34099,34104],{"type":21,"tag":3560,"props":34090,"children":34091},{},[34092],{"type":26,"value":34093},"量子回路を用いてパラメータ化された試行状態 |ψ(θ)⟩ を準備する",{"type":21,"tag":3560,"props":34095,"children":34096},{},[34097],{"type":26,"value":34098},"QPU 上で期待値 ⟨ψ(θ)|H|ψ(θ)⟩ を測定する",{"type":21,"tag":3560,"props":34100,"children":34101},{},[34102],{"type":26,"value":34103},"古典的最適化器を用いてエネルギーを最小化するように θ を更新する",{"type":21,"tag":3560,"props":34105,"children":34106},{},[34107],{"type":26,"value":34108},"収束するまで繰り返す",{"type":21,"tag":22,"props":34110,"children":34111},{},[34112],{"type":26,"value":34113},"変分原理により、任意の状態に対して ⟨ψ(θ)|H|ψ(θ)⟩ ≥ E₀ が保証されます — したがってこの量を最小化することで、真の基底状態エネルギー E₀ の上界が得られます。",{"type":21,"tag":59,"props":34115,"children":34117},{"id":34116},"セットアップ",[34118],{"type":26,"value":34116},{"type":21,"tag":22,"props":34120,"children":34121},{},[34122],{"type":26,"value":34123},"PennyLane を化学プラグインとともにインストールします:",{"type":21,"tag":312,"props":34125,"children":34127},{"code":34126,"language":10183,"meta":7,"className":10181,"style":7},"pip install pennylane pennylane-qchem\n",[34128],{"type":21,"tag":79,"props":34129,"children":34130},{"__ignoreMap":7},[34131],{"type":21,"tag":322,"props":34132,"children":34133},{"class":324,"line":325},[34134,34138,34142,34146],{"type":21,"tag":322,"props":34135,"children":34136},{"style":4801},[34137],{"type":26,"value":12232},{"type":21,"tag":322,"props":34139,"children":34140},{"style":359},[34141],{"type":26,"value":12237},{"type":21,"tag":322,"props":34143,"children":34144},{"style":359},[34145],{"type":26,"value":21037},{"type":21,"tag":322,"props":34147,"children":34148},{"style":359},[34149],{"type":26,"value":34150}," pennylane-qchem\n",{"type":21,"tag":22,"props":34152,"children":34153},{},[34154],{"type":26,"value":34155},"水素分子 (H₂) — 古典的な VQE ベンチマーク — では、2 個の電子と 4 個のスピン軌道 (4 量子ビット) が必要です:",{"type":21,"tag":312,"props":34157,"children":34159},{"code":34158,"language":3628,"meta":7,"className":3626,"style":7},"import pennylane as qml\nfrom pennylane import numpy as np\nimport pennylane.qchem as qchem\n\n# H2 at equilibrium bond length (Angstrom)\nsymbols = [\"H\", \"H\"]\ncoordinates = np.array([[0.0, 0.0, -0.6614], [0.0, 0.0, 0.6614]])\n\n# Build the qubit Hamiltonian\nH, qubits = qchem.molecular_hamiltonian(\n    symbols,\n    coordinates,\n    basis=\"sto-3g\"\n)\nprint(f\"Hamiltonian: {len(H.ops)} terms, {qubits} qubits\")\n# Hamiltonian: 15 terms, 4 qubits\n",[34160],{"type":21,"tag":79,"props":34161,"children":34162},{"__ignoreMap":7},[34163,34182,34209,34230,34237,34245,34278,34349,34356,34364,34381,34389,34397,34413,34420,34479],{"type":21,"tag":322,"props":34164,"children":34165},{"class":324,"line":325},[34166,34170,34174,34178],{"type":21,"tag":322,"props":34167,"children":34168},{"style":3638},[34169],{"type":26,"value":3651},{"type":21,"tag":322,"props":34171,"children":34172},{"style":329},[34173],{"type":26,"value":14530},{"type":21,"tag":322,"props":34175,"children":34176},{"style":3638},[34177],{"type":26,"value":4715},{"type":21,"tag":322,"props":34179,"children":34180},{"style":329},[34181],{"type":26,"value":14539},{"type":21,"tag":322,"props":34183,"children":34184},{"class":324,"line":335},[34185,34189,34193,34197,34201,34205],{"type":21,"tag":322,"props":34186,"children":34187},{"style":3638},[34188],{"type":26,"value":3641},{"type":21,"tag":322,"props":34190,"children":34191},{"style":329},[34192],{"type":26,"value":14530},{"type":21,"tag":322,"props":34194,"children":34195},{"style":3638},[34196],{"type":26,"value":3651},{"type":21,"tag":322,"props":34198,"children":34199},{"style":329},[34200],{"type":26,"value":4710},{"type":21,"tag":322,"props":34202,"children":34203},{"style":3638},[34204],{"type":26,"value":4715},{"type":21,"tag":322,"props":34206,"children":34207},{"style":329},[34208],{"type":26,"value":4720},{"type":21,"tag":322,"props":34210,"children":34211},{"class":324,"line":344},[34212,34216,34221,34225],{"type":21,"tag":322,"props":34213,"children":34214},{"style":3638},[34215],{"type":26,"value":3651},{"type":21,"tag":322,"props":34217,"children":34218},{"style":329},[34219],{"type":26,"value":34220}," pennylane.qchem ",{"type":21,"tag":322,"props":34222,"children":34223},{"style":3638},[34224],{"type":26,"value":4715},{"type":21,"tag":322,"props":34226,"children":34227},{"style":329},[34228],{"type":26,"value":34229}," qchem\n",{"type":21,"tag":322,"props":34231,"children":34232},{"class":324,"line":407},[34233],{"type":21,"tag":322,"props":34234,"children":34235},{"emptyLinePlaceholder":3662},[34236],{"type":26,"value":3665},{"type":21,"tag":322,"props":34238,"children":34239},{"class":324,"line":431},[34240],{"type":21,"tag":322,"props":34241,"children":34242},{"style":3888},[34243],{"type":26,"value":34244},"# H2 at equilibrium bond length (Angstrom)\n",{"type":21,"tag":322,"props":34246,"children":34247},{"class":324,"line":453},[34248,34253,34257,34261,34266,34270,34274],{"type":21,"tag":322,"props":34249,"children":34250},{"style":329},[34251],{"type":26,"value":34252},"symbols ",{"type":21,"tag":322,"props":34254,"children":34255},{"style":3638},[34256],{"type":26,"value":3678},{"type":21,"tag":322,"props":34258,"children":34259},{"style":329},[34260],{"type":26,"value":7098},{"type":21,"tag":322,"props":34262,"children":34263},{"style":359},[34264],{"type":26,"value":34265},"\"H\"",{"type":21,"tag":322,"props":34267,"children":34268},{"style":329},[34269],{"type":26,"value":367},{"type":21,"tag":322,"props":34271,"children":34272},{"style":359},[34273],{"type":26,"value":34265},{"type":21,"tag":322,"props":34275,"children":34276},{"style":329},[34277],{"type":26,"value":7124},{"type":21,"tag":322,"props":34279,"children":34280},{"class":324,"line":474},[34281,34286,34290,34295,34299,34303,34307,34311,34315,34320,34324,34328,34332,34336,34340,34344],{"type":21,"tag":322,"props":34282,"children":34283},{"style":329},[34284],{"type":26,"value":34285},"coordinates ",{"type":21,"tag":322,"props":34287,"children":34288},{"style":3638},[34289],{"type":26,"value":3678},{"type":21,"tag":322,"props":34291,"children":34292},{"style":329},[34293],{"type":26,"value":34294}," np.array([[",{"type":21,"tag":322,"props":34296,"children":34297},{"style":348},[34298],{"type":26,"value":15354},{"type":21,"tag":322,"props":34300,"children":34301},{"style":329},[34302],{"type":26,"value":367},{"type":21,"tag":322,"props":34304,"children":34305},{"style":348},[34306],{"type":26,"value":15354},{"type":21,"tag":322,"props":34308,"children":34309},{"style":329},[34310],{"type":26,"value":367},{"type":21,"tag":322,"props":34312,"children":34313},{"style":3638},[34314],{"type":26,"value":5285},{"type":21,"tag":322,"props":34316,"children":34317},{"style":348},[34318],{"type":26,"value":34319},"0.6614",{"type":21,"tag":322,"props":34321,"children":34322},{"style":329},[34323],{"type":26,"value":5660},{"type":21,"tag":322,"props":34325,"children":34326},{"style":348},[34327],{"type":26,"value":15354},{"type":21,"tag":322,"props":34329,"children":34330},{"style":329},[34331],{"type":26,"value":367},{"type":21,"tag":322,"props":34333,"children":34334},{"style":348},[34335],{"type":26,"value":15354},{"type":21,"tag":322,"props":34337,"children":34338},{"style":329},[34339],{"type":26,"value":367},{"type":21,"tag":322,"props":34341,"children":34342},{"style":348},[34343],{"type":26,"value":34319},{"type":21,"tag":322,"props":34345,"children":34346},{"style":329},[34347],{"type":26,"value":34348},"]])\n",{"type":21,"tag":322,"props":34350,"children":34351},{"class":324,"line":496},[34352],{"type":21,"tag":322,"props":34353,"children":34354},{"emptyLinePlaceholder":3662},[34355],{"type":26,"value":3665},{"type":21,"tag":322,"props":34357,"children":34358},{"class":324,"line":16},[34359],{"type":21,"tag":322,"props":34360,"children":34361},{"style":3888},[34362],{"type":26,"value":34363},"# Build the qubit Hamiltonian\n",{"type":21,"tag":322,"props":34365,"children":34366},{"class":324,"line":3884},[34367,34372,34376],{"type":21,"tag":322,"props":34368,"children":34369},{"style":329},[34370],{"type":26,"value":34371},"H, qubits ",{"type":21,"tag":322,"props":34373,"children":34374},{"style":3638},[34375],{"type":26,"value":3678},{"type":21,"tag":322,"props":34377,"children":34378},{"style":329},[34379],{"type":26,"value":34380}," qchem.molecular_hamiltonian(\n",{"type":21,"tag":322,"props":34382,"children":34383},{"class":324,"line":3894},[34384],{"type":21,"tag":322,"props":34385,"children":34386},{"style":329},[34387],{"type":26,"value":34388},"    symbols,\n",{"type":21,"tag":322,"props":34390,"children":34391},{"class":324,"line":1034},[34392],{"type":21,"tag":322,"props":34393,"children":34394},{"style":329},[34395],{"type":26,"value":34396},"    coordinates,\n",{"type":21,"tag":322,"props":34398,"children":34399},{"class":324,"line":3954},[34400,34404,34408],{"type":21,"tag":322,"props":34401,"children":34402},{"style":3703},[34403],{"type":26,"value":23280},{"type":21,"tag":322,"props":34405,"children":34406},{"style":3638},[34407],{"type":26,"value":3678},{"type":21,"tag":322,"props":34409,"children":34410},{"style":359},[34411],{"type":26,"value":34412},"\"sto-3g\"\n",{"type":21,"tag":322,"props":34414,"children":34415},{"class":324,"line":4002},[34416],{"type":21,"tag":322,"props":34417,"children":34418},{"style":329},[34419],{"type":26,"value":3738},{"type":21,"tag":322,"props":34421,"children":34422},{"class":324,"line":4044},[34423,34427,34431,34435,34440,34444,34449,34453,34458,34462,34466,34470,34475],{"type":21,"tag":322,"props":34424,"children":34425},{"style":348},[34426],{"type":26,"value":3753},{"type":21,"tag":322,"props":34428,"children":34429},{"style":329},[34430],{"type":26,"value":3758},{"type":21,"tag":322,"props":34432,"children":34433},{"style":3638},[34434],{"type":26,"value":3763},{"type":21,"tag":322,"props":34436,"children":34437},{"style":359},[34438],{"type":26,"value":34439},"\"Hamiltonian: ",{"type":21,"tag":322,"props":34441,"children":34442},{"style":348},[34443],{"type":26,"value":25402},{"type":21,"tag":322,"props":34445,"children":34446},{"style":329},[34447],{"type":26,"value":34448},"(H.ops)",{"type":21,"tag":322,"props":34450,"children":34451},{"style":348},[34452],{"type":26,"value":3783},{"type":21,"tag":322,"props":34454,"children":34455},{"style":359},[34456],{"type":26,"value":34457}," terms, ",{"type":21,"tag":322,"props":34459,"children":34460},{"style":348},[34461],{"type":26,"value":3773},{"type":21,"tag":322,"props":34463,"children":34464},{"style":329},[34465],{"type":26,"value":1797},{"type":21,"tag":322,"props":34467,"children":34468},{"style":348},[34469],{"type":26,"value":3783},{"type":21,"tag":322,"props":34471,"children":34472},{"style":359},[34473],{"type":26,"value":34474}," qubits\"",{"type":21,"tag":322,"props":34476,"children":34477},{"style":329},[34478],{"type":26,"value":3738},{"type":21,"tag":322,"props":34480,"children":34481},{"class":324,"line":5068},[34482],{"type":21,"tag":322,"props":34483,"children":34484},{"style":3888},[34485],{"type":26,"value":34486},"# Hamiltonian: 15 terms, 4 qubits\n",{"type":21,"tag":59,"props":34488,"children":34490},{"id":34489},"アンザッツの定義",[34491],{"type":26,"value":34489},{"type":21,"tag":22,"props":34493,"children":34494},{},[34495,34500],{"type":21,"tag":71,"props":34496,"children":34497},{},[34498],{"type":26,"value":34499},"アンザッツ",{"type":26,"value":34501}," とは、試行状態を準備するパラメータ化された回路のことです。化学問題では、UCCSD (Unitary Coupled-Cluster Singles and Doubles) アンザッツが標準的に用いられます:",{"type":21,"tag":312,"props":34503,"children":34505},{"code":34504,"language":3628,"meta":7,"className":3626,"style":7},"# Get UCCSD circuit parameters\nelectrons = 2\nsingles, doubles = qchem.excitations(electrons, qubits)\ns_wires, d_wires = qchem.excitations_to_wires(singles, doubles, wires=range(qubits))\n\n# Initial Hartree-Fock state (reference state)\nhf_state = qchem.hf_state(electrons, qubits)\n\ndev = qml.device(\"default.qubit\", wires=qubits)\n\n@qml.qnode(dev)\ndef circuit(weights, wires, s_wires=[], d_wires=[], hf_state=hf_state):\n    # Prepare HF reference state\n    qml.BasisState(hf_state, wires=wires)\n\n    # Apply UCCSD excitations\n    qml.UCCSD(weights, wires, s_wires=s_wires, d_wires=d_wires, init_state=hf_state)\n\n    return qml.expval(H)\n",[34506],{"type":21,"tag":79,"props":34507,"children":34508},{"__ignoreMap":7},[34509,34517,34533,34550,34584,34591,34599,34616,34623,34659,34666,34677,34720,34728,34749,34756,34764,34814,34821],{"type":21,"tag":322,"props":34510,"children":34511},{"class":324,"line":325},[34512],{"type":21,"tag":322,"props":34513,"children":34514},{"style":3888},[34515],{"type":26,"value":34516},"# Get UCCSD circuit parameters\n",{"type":21,"tag":322,"props":34518,"children":34519},{"class":324,"line":335},[34520,34525,34529],{"type":21,"tag":322,"props":34521,"children":34522},{"style":329},[34523],{"type":26,"value":34524},"electrons ",{"type":21,"tag":322,"props":34526,"children":34527},{"style":3638},[34528],{"type":26,"value":3678},{"type":21,"tag":322,"props":34530,"children":34531},{"style":348},[34532],{"type":26,"value":5531},{"type":21,"tag":322,"props":34534,"children":34535},{"class":324,"line":344},[34536,34541,34545],{"type":21,"tag":322,"props":34537,"children":34538},{"style":329},[34539],{"type":26,"value":34540},"singles, doubles ",{"type":21,"tag":322,"props":34542,"children":34543},{"style":3638},[34544],{"type":26,"value":3678},{"type":21,"tag":322,"props":34546,"children":34547},{"style":329},[34548],{"type":26,"value":34549}," qchem.excitations(electrons, qubits)\n",{"type":21,"tag":322,"props":34551,"children":34552},{"class":324,"line":407},[34553,34558,34562,34567,34571,34575,34579],{"type":21,"tag":322,"props":34554,"children":34555},{"style":329},[34556],{"type":26,"value":34557},"s_wires, d_wires ",{"type":21,"tag":322,"props":34559,"children":34560},{"style":3638},[34561],{"type":26,"value":3678},{"type":21,"tag":322,"props":34563,"children":34564},{"style":329},[34565],{"type":26,"value":34566}," qchem.excitations_to_wires(singles, doubles, ",{"type":21,"tag":322,"props":34568,"children":34569},{"style":3703},[34570],{"type":26,"value":14686},{"type":21,"tag":322,"props":34572,"children":34573},{"style":3638},[34574],{"type":26,"value":3678},{"type":21,"tag":322,"props":34576,"children":34577},{"style":348},[34578],{"type":26,"value":5379},{"type":21,"tag":322,"props":34580,"children":34581},{"style":329},[34582],{"type":26,"value":34583},"(qubits))\n",{"type":21,"tag":322,"props":34585,"children":34586},{"class":324,"line":431},[34587],{"type":21,"tag":322,"props":34588,"children":34589},{"emptyLinePlaceholder":3662},[34590],{"type":26,"value":3665},{"type":21,"tag":322,"props":34592,"children":34593},{"class":324,"line":453},[34594],{"type":21,"tag":322,"props":34595,"children":34596},{"style":3888},[34597],{"type":26,"value":34598},"# Initial Hartree-Fock state (reference state)\n",{"type":21,"tag":322,"props":34600,"children":34601},{"class":324,"line":474},[34602,34607,34611],{"type":21,"tag":322,"props":34603,"children":34604},{"style":329},[34605],{"type":26,"value":34606},"hf_state ",{"type":21,"tag":322,"props":34608,"children":34609},{"style":3638},[34610],{"type":26,"value":3678},{"type":21,"tag":322,"props":34612,"children":34613},{"style":329},[34614],{"type":26,"value":34615}," qchem.hf_state(electrons, qubits)\n",{"type":21,"tag":322,"props":34617,"children":34618},{"class":324,"line":496},[34619],{"type":21,"tag":322,"props":34620,"children":34621},{"emptyLinePlaceholder":3662},[34622],{"type":26,"value":3665},{"type":21,"tag":322,"props":34624,"children":34625},{"class":324,"line":16},[34626,34630,34634,34638,34642,34646,34650,34654],{"type":21,"tag":322,"props":34627,"children":34628},{"style":329},[34629],{"type":26,"value":14663},{"type":21,"tag":322,"props":34631,"children":34632},{"style":3638},[34633],{"type":26,"value":3678},{"type":21,"tag":322,"props":34635,"children":34636},{"style":329},[34637],{"type":26,"value":14672},{"type":21,"tag":322,"props":34639,"children":34640},{"style":359},[34641],{"type":26,"value":14677},{"type":21,"tag":322,"props":34643,"children":34644},{"style":329},[34645],{"type":26,"value":367},{"type":21,"tag":322,"props":34647,"children":34648},{"style":3703},[34649],{"type":26,"value":14686},{"type":21,"tag":322,"props":34651,"children":34652},{"style":3638},[34653],{"type":26,"value":3678},{"type":21,"tag":322,"props":34655,"children":34656},{"style":329},[34657],{"type":26,"value":34658},"qubits)\n",{"type":21,"tag":322,"props":34660,"children":34661},{"class":324,"line":3884},[34662],{"type":21,"tag":322,"props":34663,"children":34664},{"emptyLinePlaceholder":3662},[34665],{"type":26,"value":3665},{"type":21,"tag":322,"props":34667,"children":34668},{"class":324,"line":3894},[34669,34673],{"type":21,"tag":322,"props":34670,"children":34671},{"style":4801},[34672],{"type":26,"value":14710},{"type":21,"tag":322,"props":34674,"children":34675},{"style":329},[34676],{"type":26,"value":14715},{"type":21,"tag":322,"props":34678,"children":34679},{"class":324,"line":1034},[34680,34684,34688,34693,34697,34702,34706,34711,34715],{"type":21,"tag":322,"props":34681,"children":34682},{"style":3638},[34683],{"type":26,"value":4798},{"type":21,"tag":322,"props":34685,"children":34686},{"style":4801},[34687],{"type":26,"value":14727},{"type":21,"tag":322,"props":34689,"children":34690},{"style":329},[34691],{"type":26,"value":34692},"(weights, wires, s_wires",{"type":21,"tag":322,"props":34694,"children":34695},{"style":3638},[34696],{"type":26,"value":3678},{"type":21,"tag":322,"props":34698,"children":34699},{"style":329},[34700],{"type":26,"value":34701},"[], d_wires",{"type":21,"tag":322,"props":34703,"children":34704},{"style":3638},[34705],{"type":26,"value":3678},{"type":21,"tag":322,"props":34707,"children":34708},{"style":329},[34709],{"type":26,"value":34710},"[], hf_state",{"type":21,"tag":322,"props":34712,"children":34713},{"style":3638},[34714],{"type":26,"value":3678},{"type":21,"tag":322,"props":34716,"children":34717},{"style":329},[34718],{"type":26,"value":34719},"hf_state):\n",{"type":21,"tag":322,"props":34721,"children":34722},{"class":324,"line":3954},[34723],{"type":21,"tag":322,"props":34724,"children":34725},{"style":3888},[34726],{"type":26,"value":34727},"    # Prepare HF reference state\n",{"type":21,"tag":322,"props":34729,"children":34730},{"class":324,"line":4002},[34731,34736,34740,34744],{"type":21,"tag":322,"props":34732,"children":34733},{"style":329},[34734],{"type":26,"value":34735},"    qml.BasisState(hf_state, ",{"type":21,"tag":322,"props":34737,"children":34738},{"style":3703},[34739],{"type":26,"value":14686},{"type":21,"tag":322,"props":34741,"children":34742},{"style":3638},[34743],{"type":26,"value":3678},{"type":21,"tag":322,"props":34745,"children":34746},{"style":329},[34747],{"type":26,"value":34748},"wires)\n",{"type":21,"tag":322,"props":34750,"children":34751},{"class":324,"line":4044},[34752],{"type":21,"tag":322,"props":34753,"children":34754},{"emptyLinePlaceholder":3662},[34755],{"type":26,"value":3665},{"type":21,"tag":322,"props":34757,"children":34758},{"class":324,"line":5068},[34759],{"type":21,"tag":322,"props":34760,"children":34761},{"style":3888},[34762],{"type":26,"value":34763},"    # Apply UCCSD excitations\n",{"type":21,"tag":322,"props":34765,"children":34766},{"class":324,"line":5087},[34767,34772,34777,34781,34786,34791,34795,34800,34805,34809],{"type":21,"tag":322,"props":34768,"children":34769},{"style":329},[34770],{"type":26,"value":34771},"    qml.UCCSD(weights, wires, ",{"type":21,"tag":322,"props":34773,"children":34774},{"style":3703},[34775],{"type":26,"value":34776},"s_wires",{"type":21,"tag":322,"props":34778,"children":34779},{"style":3638},[34780],{"type":26,"value":3678},{"type":21,"tag":322,"props":34782,"children":34783},{"style":329},[34784],{"type":26,"value":34785},"s_wires, ",{"type":21,"tag":322,"props":34787,"children":34788},{"style":3703},[34789],{"type":26,"value":34790},"d_wires",{"type":21,"tag":322,"props":34792,"children":34793},{"style":3638},[34794],{"type":26,"value":3678},{"type":21,"tag":322,"props":34796,"children":34797},{"style":329},[34798],{"type":26,"value":34799},"d_wires, ",{"type":21,"tag":322,"props":34801,"children":34802},{"style":3703},[34803],{"type":26,"value":34804},"init_state",{"type":21,"tag":322,"props":34806,"children":34807},{"style":3638},[34808],{"type":26,"value":3678},{"type":21,"tag":322,"props":34810,"children":34811},{"style":329},[34812],{"type":26,"value":34813},"hf_state)\n",{"type":21,"tag":322,"props":34815,"children":34816},{"class":324,"line":5095},[34817],{"type":21,"tag":322,"props":34818,"children":34819},{"emptyLinePlaceholder":3662},[34820],{"type":26,"value":3665},{"type":21,"tag":322,"props":34822,"children":34823},{"class":324,"line":5126},[34824,34828],{"type":21,"tag":322,"props":34825,"children":34826},{"style":3638},[34827],{"type":26,"value":5023},{"type":21,"tag":322,"props":34829,"children":34830},{"style":329},[34831],{"type":26,"value":25609},{"type":21,"tag":59,"props":34833,"children":34835},{"id":34834},"vqe-最適化の実行",[34836],{"type":26,"value":34837},"VQE 最適化の実行",{"type":21,"tag":22,"props":34839,"children":34840},{},[34841],{"type":26,"value":34842},"PennyLane の自動微分を用いれば、勾配ベースの最適化器を直接使用できます:",{"type":21,"tag":312,"props":34844,"children":34846},{"code":34845,"language":3628,"meta":7,"className":3626,"style":7},"# Initial parameters (all zeros = Hartree-Fock state)\ninit_params = np.zeros(len(singles) + len(doubles), requires_grad=True)\n\n# Adam optimizer (works well for VQE)\nopt = qml.AdamOptimizer(stepsize=0.4)\n\n# Optimization loop\nenergy_history = []\nparams = init_params.copy()\n\nfor step in range(200):\n    params, energy = opt.step_and_cost(\n        lambda p: circuit(p, range(qubits), s_wires=s_wires, d_wires=d_wires),\n        params\n    )\n    energy_history.append(energy)\n\n    if step % 20 == 0:\n        print(f\"Step {step:3d}: E = {energy:.6f} Ha\")\n\nprint(f\"\\nVQE ground state energy: {energy:.6f} Ha\")\nprint(f\"Reference (exact): -1.136189 Ha\")\n",[34847],{"type":21,"tag":79,"props":34848,"children":34849},{"__ignoreMap":7},[34850,34858,34912,34919,34927,34959,34966,34974,34991,35006,35013,35044,35060,35107,35115,35122,35130,35137,35169,35235,35242,35294],{"type":21,"tag":322,"props":34851,"children":34852},{"class":324,"line":325},[34853],{"type":21,"tag":322,"props":34854,"children":34855},{"style":3888},[34856],{"type":26,"value":34857},"# Initial parameters (all zeros = Hartree-Fock state)\n",{"type":21,"tag":322,"props":34859,"children":34860},{"class":324,"line":335},[34861,34865,34869,34873,34877,34882,34886,34891,34896,34900,34904,34908],{"type":21,"tag":322,"props":34862,"children":34863},{"style":329},[34864],{"type":26,"value":30412},{"type":21,"tag":322,"props":34866,"children":34867},{"style":3638},[34868],{"type":26,"value":3678},{"type":21,"tag":322,"props":34870,"children":34871},{"style":329},[34872],{"type":26,"value":13125},{"type":21,"tag":322,"props":34874,"children":34875},{"style":348},[34876],{"type":26,"value":15502},{"type":21,"tag":322,"props":34878,"children":34879},{"style":329},[34880],{"type":26,"value":34881},"(singles) ",{"type":21,"tag":322,"props":34883,"children":34884},{"style":3638},[34885],{"type":26,"value":6965},{"type":21,"tag":322,"props":34887,"children":34888},{"style":348},[34889],{"type":26,"value":34890}," len",{"type":21,"tag":322,"props":34892,"children":34893},{"style":329},[34894],{"type":26,"value":34895},"(doubles), ",{"type":21,"tag":322,"props":34897,"children":34898},{"style":3703},[34899],{"type":26,"value":15320},{"type":21,"tag":322,"props":34901,"children":34902},{"style":3638},[34903],{"type":26,"value":3678},{"type":21,"tag":322,"props":34905,"children":34906},{"style":348},[34907],{"type":26,"value":3715},{"type":21,"tag":322,"props":34909,"children":34910},{"style":329},[34911],{"type":26,"value":3738},{"type":21,"tag":322,"props":34913,"children":34914},{"class":324,"line":344},[34915],{"type":21,"tag":322,"props":34916,"children":34917},{"emptyLinePlaceholder":3662},[34918],{"type":26,"value":3665},{"type":21,"tag":322,"props":34920,"children":34921},{"class":324,"line":407},[34922],{"type":21,"tag":322,"props":34923,"children":34924},{"style":3888},[34925],{"type":26,"value":34926},"# Adam optimizer (works well for VQE)\n",{"type":21,"tag":322,"props":34928,"children":34929},{"class":324,"line":431},[34930,34934,34938,34942,34946,34950,34955],{"type":21,"tag":322,"props":34931,"children":34932},{"style":329},[34933],{"type":26,"value":15389},{"type":21,"tag":322,"props":34935,"children":34936},{"style":3638},[34937],{"type":26,"value":3678},{"type":21,"tag":322,"props":34939,"children":34940},{"style":329},[34941],{"type":26,"value":15398},{"type":21,"tag":322,"props":34943,"children":34944},{"style":3703},[34945],{"type":26,"value":15403},{"type":21,"tag":322,"props":34947,"children":34948},{"style":3638},[34949],{"type":26,"value":3678},{"type":21,"tag":322,"props":34951,"children":34952},{"style":348},[34953],{"type":26,"value":34954},"0.4",{"type":21,"tag":322,"props":34956,"children":34957},{"style":329},[34958],{"type":26,"value":3738},{"type":21,"tag":322,"props":34960,"children":34961},{"class":324,"line":453},[34962],{"type":21,"tag":322,"props":34963,"children":34964},{"emptyLinePlaceholder":3662},[34965],{"type":26,"value":3665},{"type":21,"tag":322,"props":34967,"children":34968},{"class":324,"line":474},[34969],{"type":21,"tag":322,"props":34970,"children":34971},{"style":3888},[34972],{"type":26,"value":34973},"# Optimization loop\n",{"type":21,"tag":322,"props":34975,"children":34976},{"class":324,"line":496},[34977,34982,34986],{"type":21,"tag":322,"props":34978,"children":34979},{"style":329},[34980],{"type":26,"value":34981},"energy_history ",{"type":21,"tag":322,"props":34983,"children":34984},{"style":3638},[34985],{"type":26,"value":3678},{"type":21,"tag":322,"props":34987,"children":34988},{"style":329},[34989],{"type":26,"value":34990}," []\n",{"type":21,"tag":322,"props":34992,"children":34993},{"class":324,"line":16},[34994,34998,35002],{"type":21,"tag":322,"props":34995,"children":34996},{"style":329},[34997],{"type":26,"value":25672},{"type":21,"tag":322,"props":34999,"children":35000},{"style":3638},[35001],{"type":26,"value":3678},{"type":21,"tag":322,"props":35003,"children":35004},{"style":329},[35005],{"type":26,"value":26341},{"type":21,"tag":322,"props":35007,"children":35008},{"class":324,"line":3884},[35009],{"type":21,"tag":322,"props":35010,"children":35011},{"emptyLinePlaceholder":3662},[35012],{"type":26,"value":3665},{"type":21,"tag":322,"props":35014,"children":35015},{"class":324,"line":3894},[35016,35020,35024,35028,35032,35036,35040],{"type":21,"tag":322,"props":35017,"children":35018},{"style":3638},[35019],{"type":26,"value":3931},{"type":21,"tag":322,"props":35021,"children":35022},{"style":329},[35023],{"type":26,"value":15451},{"type":21,"tag":322,"props":35025,"children":35026},{"style":3638},[35027],{"type":26,"value":3941},{"type":21,"tag":322,"props":35029,"children":35030},{"style":348},[35031],{"type":26,"value":3946},{"type":21,"tag":322,"props":35033,"children":35034},{"style":329},[35035],{"type":26,"value":3758},{"type":21,"tag":322,"props":35037,"children":35038},{"style":348},[35039],{"type":26,"value":15068},{"type":21,"tag":322,"props":35041,"children":35042},{"style":329},[35043],{"type":26,"value":4851},{"type":21,"tag":322,"props":35045,"children":35046},{"class":324,"line":1034},[35047,35051,35055],{"type":21,"tag":322,"props":35048,"children":35049},{"style":329},[35050],{"type":26,"value":26410},{"type":21,"tag":322,"props":35052,"children":35053},{"style":3638},[35054],{"type":26,"value":3678},{"type":21,"tag":322,"props":35056,"children":35057},{"style":329},[35058],{"type":26,"value":35059}," opt.step_and_cost(\n",{"type":21,"tag":322,"props":35061,"children":35062},{"class":324,"line":3954},[35063,35068,35073,35077,35082,35086,35090,35094,35098,35102],{"type":21,"tag":322,"props":35064,"children":35065},{"style":3638},[35066],{"type":26,"value":35067},"        lambda",{"type":21,"tag":322,"props":35069,"children":35070},{"style":329},[35071],{"type":26,"value":35072}," p: circuit(p, ",{"type":21,"tag":322,"props":35074,"children":35075},{"style":348},[35076],{"type":26,"value":5379},{"type":21,"tag":322,"props":35078,"children":35079},{"style":329},[35080],{"type":26,"value":35081},"(qubits), ",{"type":21,"tag":322,"props":35083,"children":35084},{"style":3703},[35085],{"type":26,"value":34776},{"type":21,"tag":322,"props":35087,"children":35088},{"style":3638},[35089],{"type":26,"value":3678},{"type":21,"tag":322,"props":35091,"children":35092},{"style":329},[35093],{"type":26,"value":34785},{"type":21,"tag":322,"props":35095,"children":35096},{"style":3703},[35097],{"type":26,"value":34790},{"type":21,"tag":322,"props":35099,"children":35100},{"style":3638},[35101],{"type":26,"value":3678},{"type":21,"tag":322,"props":35103,"children":35104},{"style":329},[35105],{"type":26,"value":35106},"d_wires),\n",{"type":21,"tag":322,"props":35108,"children":35109},{"class":324,"line":4002},[35110],{"type":21,"tag":322,"props":35111,"children":35112},{"style":329},[35113],{"type":26,"value":35114},"        params\n",{"type":21,"tag":322,"props":35116,"children":35117},{"class":324,"line":4044},[35118],{"type":21,"tag":322,"props":35119,"children":35120},{"style":329},[35121],{"type":26,"value":5015},{"type":21,"tag":322,"props":35123,"children":35124},{"class":324,"line":5068},[35125],{"type":21,"tag":322,"props":35126,"children":35127},{"style":329},[35128],{"type":26,"value":35129},"    energy_history.append(energy)\n",{"type":21,"tag":322,"props":35131,"children":35132},{"class":324,"line":5087},[35133],{"type":21,"tag":322,"props":35134,"children":35135},{"emptyLinePlaceholder":3662},[35136],{"type":26,"value":3665},{"type":21,"tag":322,"props":35138,"children":35139},{"class":324,"line":5095},[35140,35144,35148,35152,35157,35161,35165],{"type":21,"tag":322,"props":35141,"children":35142},{"style":3638},[35143],{"type":26,"value":12384},{"type":21,"tag":322,"props":35145,"children":35146},{"style":329},[35147],{"type":26,"value":15451},{"type":21,"tag":322,"props":35149,"children":35150},{"style":3638},[35151],{"type":26,"value":6816},{"type":21,"tag":322,"props":35153,"children":35154},{"style":348},[35155],{"type":26,"value":35156}," 20",{"type":21,"tag":322,"props":35158,"children":35159},{"style":3638},[35160],{"type":26,"value":15577},{"type":21,"tag":322,"props":35162,"children":35163},{"style":348},[35164],{"type":26,"value":12398},{"type":21,"tag":322,"props":35166,"children":35167},{"style":329},[35168],{"type":26,"value":5356},{"type":21,"tag":322,"props":35170,"children":35171},{"class":324,"line":5126},[35172,35176,35180,35184,35188,35192,35196,35201,35205,35210,35214,35218,35223,35227,35231],{"type":21,"tag":322,"props":35173,"children":35174},{"style":348},[35175],{"type":26,"value":13330},{"type":21,"tag":322,"props":35177,"children":35178},{"style":329},[35179],{"type":26,"value":3758},{"type":21,"tag":322,"props":35181,"children":35182},{"style":3638},[35183],{"type":26,"value":3763},{"type":21,"tag":322,"props":35185,"children":35186},{"style":359},[35187],{"type":26,"value":15668},{"type":21,"tag":322,"props":35189,"children":35190},{"style":348},[35191],{"type":26,"value":3773},{"type":21,"tag":322,"props":35193,"children":35194},{"style":329},[35195],{"type":26,"value":15677},{"type":21,"tag":322,"props":35197,"children":35198},{"style":3638},[35199],{"type":26,"value":35200},":3d",{"type":21,"tag":322,"props":35202,"children":35203},{"style":348},[35204],{"type":26,"value":3783},{"type":21,"tag":322,"props":35206,"children":35207},{"style":359},[35208],{"type":26,"value":35209},": E = ",{"type":21,"tag":322,"props":35211,"children":35212},{"style":348},[35213],{"type":26,"value":3773},{"type":21,"tag":322,"props":35215,"children":35216},{"style":329},[35217],{"type":26,"value":27314},{"type":21,"tag":322,"props":35219,"children":35220},{"style":3638},[35221],{"type":26,"value":35222},":.6f",{"type":21,"tag":322,"props":35224,"children":35225},{"style":348},[35226],{"type":26,"value":3783},{"type":21,"tag":322,"props":35228,"children":35229},{"style":359},[35230],{"type":26,"value":19868},{"type":21,"tag":322,"props":35232,"children":35233},{"style":329},[35234],{"type":26,"value":3738},{"type":21,"tag":322,"props":35236,"children":35237},{"class":324,"line":5171},[35238],{"type":21,"tag":322,"props":35239,"children":35240},{"emptyLinePlaceholder":3662},[35241],{"type":26,"value":3665},{"type":21,"tag":322,"props":35243,"children":35244},{"class":324,"line":5210},[35245,35249,35253,35257,35261,35265,35270,35274,35278,35282,35286,35290],{"type":21,"tag":322,"props":35246,"children":35247},{"style":348},[35248],{"type":26,"value":3753},{"type":21,"tag":322,"props":35250,"children":35251},{"style":329},[35252],{"type":26,"value":3758},{"type":21,"tag":322,"props":35254,"children":35255},{"style":3638},[35256],{"type":26,"value":3763},{"type":21,"tag":322,"props":35258,"children":35259},{"style":359},[35260],{"type":26,"value":3788},{"type":21,"tag":322,"props":35262,"children":35263},{"style":348},[35264],{"type":26,"value":15805},{"type":21,"tag":322,"props":35266,"children":35267},{"style":359},[35268],{"type":26,"value":35269},"VQE ground state energy: ",{"type":21,"tag":322,"props":35271,"children":35272},{"style":348},[35273],{"type":26,"value":3773},{"type":21,"tag":322,"props":35275,"children":35276},{"style":329},[35277],{"type":26,"value":27314},{"type":21,"tag":322,"props":35279,"children":35280},{"style":3638},[35281],{"type":26,"value":35222},{"type":21,"tag":322,"props":35283,"children":35284},{"style":348},[35285],{"type":26,"value":3783},{"type":21,"tag":322,"props":35287,"children":35288},{"style":359},[35289],{"type":26,"value":19868},{"type":21,"tag":322,"props":35291,"children":35292},{"style":329},[35293],{"type":26,"value":3738},{"type":21,"tag":322,"props":35295,"children":35296},{"class":324,"line":5228},[35297,35301,35305,35309,35314],{"type":21,"tag":322,"props":35298,"children":35299},{"style":348},[35300],{"type":26,"value":3753},{"type":21,"tag":322,"props":35302,"children":35303},{"style":329},[35304],{"type":26,"value":3758},{"type":21,"tag":322,"props":35306,"children":35307},{"style":3638},[35308],{"type":26,"value":3763},{"type":21,"tag":322,"props":35310,"children":35311},{"style":359},[35312],{"type":26,"value":35313},"\"Reference (exact): -1.136189 Ha\"",{"type":21,"tag":322,"props":35315,"children":35316},{"style":329},[35317],{"type":26,"value":3738},{"type":21,"tag":22,"props":35319,"children":35320},{},[35321],{"type":26,"value":35322},"典型的な出力:",{"type":21,"tag":312,"props":35324,"children":35326},{"code":35325},"Step   0: E = -1.117498 Ha\nStep  20: E = -1.133254 Ha\nStep  40: E = -1.135901 Ha\nStep  60: E = -1.136140 Ha\n...\nVQE ground state energy: -1.136174 Ha\nReference (exact): -1.136189 Ha\n",[35327],{"type":21,"tag":79,"props":35328,"children":35329},{"__ignoreMap":7},[35330],{"type":26,"value":35325},{"type":21,"tag":22,"props":35332,"children":35333},{},[35334],{"type":26,"value":35335},"VQE は正確なエネルギーの約 0.015 mHa 以内に到達します — H₂ における化学的精度です。",{"type":21,"tag":59,"props":35337,"children":35339},{"id":35338},"勾配フリー最適化器の使用",[35340],{"type":26,"value":35338},{"type":21,"tag":22,"props":35342,"children":35343},{},[35344],{"type":26,"value":35345},"ノイズのあるハードウェアでは、ハードウェアの勾配がノイジーであるため、COBYLA や SPSA のような勾配フリー最適化器がしばしば有利です:",{"type":21,"tag":312,"props":35347,"children":35349},{"code":35348,"language":3628,"meta":7,"className":3626,"style":7},"from scipy.optimize import minimize\n\n# Objective function (no gradient needed)\ndef objective(params):\n    return float(circuit(params, range(qubits), s_wires=s_wires, d_wires=d_wires))\n\nresult = minimize(\n    objective,\n    x0=init_params,\n    method=\"COBYLA\",\n    options={\"maxiter\": 500, \"rhobeg\": 0.1}\n)\n\nprint(f\"COBYLA energy: {result.fun:.6f} Ha\")\n",[35350],{"type":21,"tag":79,"props":35351,"children":35352},{"__ignoreMap":7},[35353,35372,35379,35387,35403,35452,35459,35475,35483,35500,35521,35572,35579,35586],{"type":21,"tag":322,"props":35354,"children":35355},{"class":324,"line":325},[35356,35360,35364,35368],{"type":21,"tag":322,"props":35357,"children":35358},{"style":3638},[35359],{"type":26,"value":3641},{"type":21,"tag":322,"props":35361,"children":35362},{"style":329},[35363],{"type":26,"value":29525},{"type":21,"tag":322,"props":35365,"children":35366},{"style":3638},[35367],{"type":26,"value":3651},{"type":21,"tag":322,"props":35369,"children":35370},{"style":329},[35371],{"type":26,"value":29534},{"type":21,"tag":322,"props":35373,"children":35374},{"class":324,"line":335},[35375],{"type":21,"tag":322,"props":35376,"children":35377},{"emptyLinePlaceholder":3662},[35378],{"type":26,"value":3665},{"type":21,"tag":322,"props":35380,"children":35381},{"class":324,"line":344},[35382],{"type":21,"tag":322,"props":35383,"children":35384},{"style":3888},[35385],{"type":26,"value":35386},"# Objective function (no gradient needed)\n",{"type":21,"tag":322,"props":35388,"children":35389},{"class":324,"line":407},[35390,35394,35399],{"type":21,"tag":322,"props":35391,"children":35392},{"style":3638},[35393],{"type":26,"value":4798},{"type":21,"tag":322,"props":35395,"children":35396},{"style":4801},[35397],{"type":26,"value":35398}," objective",{"type":21,"tag":322,"props":35400,"children":35401},{"style":329},[35402],{"type":26,"value":25581},{"type":21,"tag":322,"props":35404,"children":35405},{"class":324,"line":431},[35406,35410,35414,35419,35423,35427,35431,35435,35439,35443,35447],{"type":21,"tag":322,"props":35407,"children":35408},{"style":3638},[35409],{"type":26,"value":5023},{"type":21,"tag":322,"props":35411,"children":35412},{"style":348},[35413],{"type":26,"value":26358},{"type":21,"tag":322,"props":35415,"children":35416},{"style":329},[35417],{"type":26,"value":35418},"(circuit(params, ",{"type":21,"tag":322,"props":35420,"children":35421},{"style":348},[35422],{"type":26,"value":5379},{"type":21,"tag":322,"props":35424,"children":35425},{"style":329},[35426],{"type":26,"value":35081},{"type":21,"tag":322,"props":35428,"children":35429},{"style":3703},[35430],{"type":26,"value":34776},{"type":21,"tag":322,"props":35432,"children":35433},{"style":3638},[35434],{"type":26,"value":3678},{"type":21,"tag":322,"props":35436,"children":35437},{"style":329},[35438],{"type":26,"value":34785},{"type":21,"tag":322,"props":35440,"children":35441},{"style":3703},[35442],{"type":26,"value":34790},{"type":21,"tag":322,"props":35444,"children":35445},{"style":3638},[35446],{"type":26,"value":3678},{"type":21,"tag":322,"props":35448,"children":35449},{"style":329},[35450],{"type":26,"value":35451},"d_wires))\n",{"type":21,"tag":322,"props":35453,"children":35454},{"class":324,"line":453},[35455],{"type":21,"tag":322,"props":35456,"children":35457},{"emptyLinePlaceholder":3662},[35458],{"type":26,"value":3665},{"type":21,"tag":322,"props":35460,"children":35461},{"class":324,"line":474},[35462,35466,35470],{"type":21,"tag":322,"props":35463,"children":35464},{"style":329},[35465],{"type":26,"value":19764},{"type":21,"tag":322,"props":35467,"children":35468},{"style":3638},[35469],{"type":26,"value":3678},{"type":21,"tag":322,"props":35471,"children":35472},{"style":329},[35473],{"type":26,"value":35474}," minimize(\n",{"type":21,"tag":322,"props":35476,"children":35477},{"class":324,"line":496},[35478],{"type":21,"tag":322,"props":35479,"children":35480},{"style":329},[35481],{"type":26,"value":35482},"    objective,\n",{"type":21,"tag":322,"props":35484,"children":35485},{"class":324,"line":16},[35486,35491,35495],{"type":21,"tag":322,"props":35487,"children":35488},{"style":3703},[35489],{"type":26,"value":35490},"    x0",{"type":21,"tag":322,"props":35492,"children":35493},{"style":3638},[35494],{"type":26,"value":3678},{"type":21,"tag":322,"props":35496,"children":35497},{"style":329},[35498],{"type":26,"value":35499},"init_params,\n",{"type":21,"tag":322,"props":35501,"children":35502},{"class":324,"line":3884},[35503,35508,35512,35517],{"type":21,"tag":322,"props":35504,"children":35505},{"style":3703},[35506],{"type":26,"value":35507},"    method",{"type":21,"tag":322,"props":35509,"children":35510},{"style":3638},[35511],{"type":26,"value":3678},{"type":21,"tag":322,"props":35513,"children":35514},{"style":359},[35515],{"type":26,"value":35516},"\"COBYLA\"",{"type":21,"tag":322,"props":35518,"children":35519},{"style":329},[35520],{"type":26,"value":428},{"type":21,"tag":322,"props":35522,"children":35523},{"class":324,"line":3894},[35524,35529,35533,35537,35542,35546,35551,35555,35560,35564,35568],{"type":21,"tag":322,"props":35525,"children":35526},{"style":3703},[35527],{"type":26,"value":35528},"    options",{"type":21,"tag":322,"props":35530,"children":35531},{"style":3638},[35532],{"type":26,"value":3678},{"type":21,"tag":322,"props":35534,"children":35535},{"style":329},[35536],{"type":26,"value":3773},{"type":21,"tag":322,"props":35538,"children":35539},{"style":359},[35540],{"type":26,"value":35541},"\"maxiter\"",{"type":21,"tag":322,"props":35543,"children":35544},{"style":329},[35545],{"type":26,"value":418},{"type":21,"tag":322,"props":35547,"children":35548},{"style":348},[35549],{"type":26,"value":35550},"500",{"type":21,"tag":322,"props":35552,"children":35553},{"style":329},[35554],{"type":26,"value":367},{"type":21,"tag":322,"props":35556,"children":35557},{"style":359},[35558],{"type":26,"value":35559},"\"rhobeg\"",{"type":21,"tag":322,"props":35561,"children":35562},{"style":329},[35563],{"type":26,"value":418},{"type":21,"tag":322,"props":35565,"children":35566},{"style":348},[35567],{"type":26,"value":15310},{"type":21,"tag":322,"props":35569,"children":35570},{"style":329},[35571],{"type":26,"value":519},{"type":21,"tag":322,"props":35573,"children":35574},{"class":324,"line":1034},[35575],{"type":21,"tag":322,"props":35576,"children":35577},{"style":329},[35578],{"type":26,"value":3738},{"type":21,"tag":322,"props":35580,"children":35581},{"class":324,"line":3954},[35582],{"type":21,"tag":322,"props":35583,"children":35584},{"emptyLinePlaceholder":3662},[35585],{"type":26,"value":3665},{"type":21,"tag":322,"props":35587,"children":35588},{"class":324,"line":4002},[35589,35593,35597,35601,35606,35610,35614,35618,35622,35626],{"type":21,"tag":322,"props":35590,"children":35591},{"style":348},[35592],{"type":26,"value":3753},{"type":21,"tag":322,"props":35594,"children":35595},{"style":329},[35596],{"type":26,"value":3758},{"type":21,"tag":322,"props":35598,"children":35599},{"style":3638},[35600],{"type":26,"value":3763},{"type":21,"tag":322,"props":35602,"children":35603},{"style":359},[35604],{"type":26,"value":35605},"\"COBYLA energy: ",{"type":21,"tag":322,"props":35607,"children":35608},{"style":348},[35609],{"type":26,"value":3773},{"type":21,"tag":322,"props":35611,"children":35612},{"style":329},[35613],{"type":26,"value":30590},{"type":21,"tag":322,"props":35615,"children":35616},{"style":3638},[35617],{"type":26,"value":35222},{"type":21,"tag":322,"props":35619,"children":35620},{"style":348},[35621],{"type":26,"value":3783},{"type":21,"tag":322,"props":35623,"children":35624},{"style":359},[35625],{"type":26,"value":19868},{"type":21,"tag":322,"props":35627,"children":35628},{"style":329},[35629],{"type":26,"value":3738},{"type":21,"tag":59,"props":35631,"children":35633},{"id":35632},"hlquantum-で-vqe-を実行する",[35634],{"type":26,"value":35635},"HLQuantum で VQE を実行する",{"type":21,"tag":22,"props":35637,"children":35638},{},[35639],{"type":26,"value":35640},"HLQuantum には、すべてのバックエンドで動作する組み込みの VQE 実装が含まれています:",{"type":21,"tag":312,"props":35642,"children":35644},{"code":35643,"language":3628,"meta":7,"className":3626,"style":7},"import hlquantum as hlq\n\n# Define the Hamiltonian in HLQuantum's format\nH = hlq.hamiltonians.h2_molecule(bond_length=1.32)\n\n# Run VQE on any backend\nresult = hlq.algorithms.vqe(\n    hamiltonian=H,\n    ansatz=\"uccsd\",\n    optimizer=\"adam\",\n    max_iterations=200,\n    backend=\"pennylane\",   # or \"qiskit\", \"cudaq\"\n)\n\nprint(f\"Ground state energy: {result.energy:.6f} Ha\")\nprint(f\"Optimal parameters: {result.params}\")\nprint(f\"Converged in {result.iterations} iterations\")\n",[35645],{"type":21,"tag":79,"props":35646,"children":35647},{"__ignoreMap":7},[35648,35667,35674,35682,35716,35723,35731,35746,35763,35784,35804,35824,35849,35856,35863,35906,35947],{"type":21,"tag":322,"props":35649,"children":35650},{"class":324,"line":325},[35651,35655,35659,35663],{"type":21,"tag":322,"props":35652,"children":35653},{"style":3638},[35654],{"type":26,"value":3651},{"type":21,"tag":322,"props":35656,"children":35657},{"style":329},[35658],{"type":26,"value":19732},{"type":21,"tag":322,"props":35660,"children":35661},{"style":3638},[35662],{"type":26,"value":4715},{"type":21,"tag":322,"props":35664,"children":35665},{"style":329},[35666],{"type":26,"value":19741},{"type":21,"tag":322,"props":35668,"children":35669},{"class":324,"line":335},[35670],{"type":21,"tag":322,"props":35671,"children":35672},{"emptyLinePlaceholder":3662},[35673],{"type":26,"value":3665},{"type":21,"tag":322,"props":35675,"children":35676},{"class":324,"line":344},[35677],{"type":21,"tag":322,"props":35678,"children":35679},{"style":3888},[35680],{"type":26,"value":35681},"# Define the Hamiltonian in HLQuantum's format\n",{"type":21,"tag":322,"props":35683,"children":35684},{"class":324,"line":407},[35685,35689,35693,35698,35703,35707,35712],{"type":21,"tag":322,"props":35686,"children":35687},{"style":329},[35688],{"type":26,"value":25318},{"type":21,"tag":322,"props":35690,"children":35691},{"style":3638},[35692],{"type":26,"value":3678},{"type":21,"tag":322,"props":35694,"children":35695},{"style":329},[35696],{"type":26,"value":35697}," hlq.hamiltonians.h2_molecule(",{"type":21,"tag":322,"props":35699,"children":35700},{"style":3703},[35701],{"type":26,"value":35702},"bond_length",{"type":21,"tag":322,"props":35704,"children":35705},{"style":3638},[35706],{"type":26,"value":3678},{"type":21,"tag":322,"props":35708,"children":35709},{"style":348},[35710],{"type":26,"value":35711},"1.32",{"type":21,"tag":322,"props":35713,"children":35714},{"style":329},[35715],{"type":26,"value":3738},{"type":21,"tag":322,"props":35717,"children":35718},{"class":324,"line":431},[35719],{"type":21,"tag":322,"props":35720,"children":35721},{"emptyLinePlaceholder":3662},[35722],{"type":26,"value":3665},{"type":21,"tag":322,"props":35724,"children":35725},{"class":324,"line":453},[35726],{"type":21,"tag":322,"props":35727,"children":35728},{"style":3888},[35729],{"type":26,"value":35730},"# Run VQE on any backend\n",{"type":21,"tag":322,"props":35732,"children":35733},{"class":324,"line":474},[35734,35738,35742],{"type":21,"tag":322,"props":35735,"children":35736},{"style":329},[35737],{"type":26,"value":19764},{"type":21,"tag":322,"props":35739,"children":35740},{"style":3638},[35741],{"type":26,"value":3678},{"type":21,"tag":322,"props":35743,"children":35744},{"style":329},[35745],{"type":26,"value":23252},{"type":21,"tag":322,"props":35747,"children":35748},{"class":324,"line":496},[35749,35754,35758],{"type":21,"tag":322,"props":35750,"children":35751},{"style":3703},[35752],{"type":26,"value":35753},"    hamiltonian",{"type":21,"tag":322,"props":35755,"children":35756},{"style":3638},[35757],{"type":26,"value":3678},{"type":21,"tag":322,"props":35759,"children":35760},{"style":329},[35761],{"type":26,"value":35762},"H,\n",{"type":21,"tag":322,"props":35764,"children":35765},{"class":324,"line":16},[35766,35771,35775,35780],{"type":21,"tag":322,"props":35767,"children":35768},{"style":3703},[35769],{"type":26,"value":35770},"    ansatz",{"type":21,"tag":322,"props":35772,"children":35773},{"style":3638},[35774],{"type":26,"value":3678},{"type":21,"tag":322,"props":35776,"children":35777},{"style":359},[35778],{"type":26,"value":35779},"\"uccsd\"",{"type":21,"tag":322,"props":35781,"children":35782},{"style":329},[35783],{"type":26,"value":428},{"type":21,"tag":322,"props":35785,"children":35786},{"class":324,"line":3884},[35787,35791,35795,35800],{"type":21,"tag":322,"props":35788,"children":35789},{"style":3703},[35790],{"type":26,"value":23320},{"type":21,"tag":322,"props":35792,"children":35793},{"style":3638},[35794],{"type":26,"value":3678},{"type":21,"tag":322,"props":35796,"children":35797},{"style":359},[35798],{"type":26,"value":35799},"\"adam\"",{"type":21,"tag":322,"props":35801,"children":35802},{"style":329},[35803],{"type":26,"value":428},{"type":21,"tag":322,"props":35805,"children":35806},{"class":324,"line":3894},[35807,35812,35816,35820],{"type":21,"tag":322,"props":35808,"children":35809},{"style":3703},[35810],{"type":26,"value":35811},"    max_iterations",{"type":21,"tag":322,"props":35813,"children":35814},{"style":3638},[35815],{"type":26,"value":3678},{"type":21,"tag":322,"props":35817,"children":35818},{"style":348},[35819],{"type":26,"value":15068},{"type":21,"tag":322,"props":35821,"children":35822},{"style":329},[35823],{"type":26,"value":428},{"type":21,"tag":322,"props":35825,"children":35826},{"class":324,"line":1034},[35827,35831,35835,35839,35844],{"type":21,"tag":322,"props":35828,"children":35829},{"style":3703},[35830],{"type":26,"value":23300},{"type":21,"tag":322,"props":35832,"children":35833},{"style":3638},[35834],{"type":26,"value":3678},{"type":21,"tag":322,"props":35836,"children":35837},{"style":359},[35838],{"type":26,"value":33895},{"type":21,"tag":322,"props":35840,"children":35841},{"style":329},[35842],{"type":26,"value":35843},",   ",{"type":21,"tag":322,"props":35845,"children":35846},{"style":3888},[35847],{"type":26,"value":35848},"# or \"qiskit\", \"cudaq\"\n",{"type":21,"tag":322,"props":35850,"children":35851},{"class":324,"line":3954},[35852],{"type":21,"tag":322,"props":35853,"children":35854},{"style":329},[35855],{"type":26,"value":3738},{"type":21,"tag":322,"props":35857,"children":35858},{"class":324,"line":4002},[35859],{"type":21,"tag":322,"props":35860,"children":35861},{"emptyLinePlaceholder":3662},[35862],{"type":26,"value":3665},{"type":21,"tag":322,"props":35864,"children":35865},{"class":324,"line":4044},[35866,35870,35874,35878,35882,35886,35890,35894,35898,35902],{"type":21,"tag":322,"props":35867,"children":35868},{"style":348},[35869],{"type":26,"value":3753},{"type":21,"tag":322,"props":35871,"children":35872},{"style":329},[35873],{"type":26,"value":3758},{"type":21,"tag":322,"props":35875,"children":35876},{"style":3638},[35877],{"type":26,"value":3763},{"type":21,"tag":322,"props":35879,"children":35880},{"style":359},[35881],{"type":26,"value":27305},{"type":21,"tag":322,"props":35883,"children":35884},{"style":348},[35885],{"type":26,"value":3773},{"type":21,"tag":322,"props":35887,"children":35888},{"style":329},[35889],{"type":26,"value":19855},{"type":21,"tag":322,"props":35891,"children":35892},{"style":3638},[35893],{"type":26,"value":35222},{"type":21,"tag":322,"props":35895,"children":35896},{"style":348},[35897],{"type":26,"value":3783},{"type":21,"tag":322,"props":35899,"children":35900},{"style":359},[35901],{"type":26,"value":19868},{"type":21,"tag":322,"props":35903,"children":35904},{"style":329},[35905],{"type":26,"value":3738},{"type":21,"tag":322,"props":35907,"children":35908},{"class":324,"line":5068},[35909,35913,35917,35921,35926,35930,35935,35939,35943],{"type":21,"tag":322,"props":35910,"children":35911},{"style":348},[35912],{"type":26,"value":3753},{"type":21,"tag":322,"props":35914,"children":35915},{"style":329},[35916],{"type":26,"value":3758},{"type":21,"tag":322,"props":35918,"children":35919},{"style":3638},[35920],{"type":26,"value":3763},{"type":21,"tag":322,"props":35922,"children":35923},{"style":359},[35924],{"type":26,"value":35925},"\"Optimal parameters: ",{"type":21,"tag":322,"props":35927,"children":35928},{"style":348},[35929],{"type":26,"value":3773},{"type":21,"tag":322,"props":35931,"children":35932},{"style":329},[35933],{"type":26,"value":35934},"result.params",{"type":21,"tag":322,"props":35936,"children":35937},{"style":348},[35938],{"type":26,"value":3783},{"type":21,"tag":322,"props":35940,"children":35941},{"style":359},[35942],{"type":26,"value":3788},{"type":21,"tag":322,"props":35944,"children":35945},{"style":329},[35946],{"type":26,"value":3738},{"type":21,"tag":322,"props":35948,"children":35949},{"class":324,"line":5087},[35950,35954,35958,35962,35967,35971,35976,35980,35985],{"type":21,"tag":322,"props":35951,"children":35952},{"style":348},[35953],{"type":26,"value":3753},{"type":21,"tag":322,"props":35955,"children":35956},{"style":329},[35957],{"type":26,"value":3758},{"type":21,"tag":322,"props":35959,"children":35960},{"style":3638},[35961],{"type":26,"value":3763},{"type":21,"tag":322,"props":35963,"children":35964},{"style":359},[35965],{"type":26,"value":35966},"\"Converged in ",{"type":21,"tag":322,"props":35968,"children":35969},{"style":348},[35970],{"type":26,"value":3773},{"type":21,"tag":322,"props":35972,"children":35973},{"style":329},[35974],{"type":26,"value":35975},"result.iterations",{"type":21,"tag":322,"props":35977,"children":35978},{"style":348},[35979],{"type":26,"value":3783},{"type":21,"tag":322,"props":35981,"children":35982},{"style":359},[35983],{"type":26,"value":35984}," iterations\"",{"type":21,"tag":322,"props":35986,"children":35987},{"style":329},[35988],{"type":26,"value":3738},{"type":21,"tag":59,"props":35990,"children":35992},{"id":35991},"実機ハードウェア向けのヒント",[35993],{"type":26,"value":35991},{"type":21,"tag":22,"props":35995,"children":35996},{},[35997],{"type":26,"value":35998},"実際の QPU (IBM Quantum、IonQ) 上で VQE を実行する際には、いくつかの追加的な考慮事項があります:",{"type":21,"tag":22,"props":36000,"children":36001},{},[36002,36007],{"type":21,"tag":71,"props":36003,"children":36004},{},[36005],{"type":26,"value":36006},"ステップあたりのショット数を減らす。",{"type":26,"value":36008}," 最適化ステップあたり 1000 ショットあれば、通常は勾配推定に十分です。毎ステップで 10,000 ショットを使うのはやめましょう — QPU の時間を浪費します。",{"type":21,"tag":22,"props":36010,"children":36011},{},[36012,36017],{"type":21,"tag":71,"props":36013,"children":36014},{},[36015],{"type":26,"value":36016},"浅い回路から始める。",{"type":26,"value":36018}," CNOT ゲートが少ないほど、ノイズも少なくなります。ハードウェアでは、UCCSD ではなくハードウェア効率的なアンザッツ回路を検討しましょう。",{"type":21,"tag":22,"props":36020,"children":36021},{},[36022,36027,36029,36035],{"type":21,"tag":71,"props":36023,"children":36024},{},[36025],{"type":26,"value":36026},"エラー緩和を有効にする。",{"type":26,"value":36028}," HLQuantum の ",{"type":21,"tag":79,"props":36030,"children":36032},{"className":36031},[],[36033],{"type":26,"value":36034},"error_mitigation=\"zne\"",{"type":26,"value":36036}," はゼロノイズ外挿 (Zero Noise Extrapolation) を適用し、ノイズのあるハードウェア上で結果を大幅に改善できます:",{"type":21,"tag":312,"props":36038,"children":36040},{"code":36039,"language":3628,"meta":7,"className":3626,"style":7},"result = hlq.run(vqe_circuit, backend=\"qiskit\", device=\"ibm_sherbrooke\",\n                 error_mitigation=\"zne\", shots=2048)\n",[36041],{"type":21,"tag":79,"props":36042,"children":36043},{"__ignoreMap":7},[36044,36092],{"type":21,"tag":322,"props":36045,"children":36046},{"class":324,"line":325},[36047,36051,36055,36060,36064,36068,36072,36076,36080,36084,36088],{"type":21,"tag":322,"props":36048,"children":36049},{"style":329},[36050],{"type":26,"value":19764},{"type":21,"tag":322,"props":36052,"children":36053},{"style":3638},[36054],{"type":26,"value":3678},{"type":21,"tag":322,"props":36056,"children":36057},{"style":329},[36058],{"type":26,"value":36059}," hlq.run(vqe_circuit, ",{"type":21,"tag":322,"props":36061,"children":36062},{"style":3703},[36063],{"type":26,"value":18908},{"type":21,"tag":322,"props":36065,"children":36066},{"style":3638},[36067],{"type":26,"value":3678},{"type":21,"tag":322,"props":36069,"children":36070},{"style":359},[36071],{"type":26,"value":29206},{"type":21,"tag":322,"props":36073,"children":36074},{"style":329},[36075],{"type":26,"value":367},{"type":21,"tag":322,"props":36077,"children":36078},{"style":3703},[36079],{"type":26,"value":22723},{"type":21,"tag":322,"props":36081,"children":36082},{"style":3638},[36083],{"type":26,"value":3678},{"type":21,"tag":322,"props":36085,"children":36086},{"style":359},[36087],{"type":26,"value":29227},{"type":21,"tag":322,"props":36089,"children":36090},{"style":329},[36091],{"type":26,"value":428},{"type":21,"tag":322,"props":36093,"children":36094},{"class":324,"line":335},[36095,36100,36104,36109,36113,36117,36121,36126],{"type":21,"tag":322,"props":36096,"children":36097},{"style":3703},[36098],{"type":26,"value":36099},"                 error_mitigation",{"type":21,"tag":322,"props":36101,"children":36102},{"style":3638},[36103],{"type":26,"value":3678},{"type":21,"tag":322,"props":36105,"children":36106},{"style":359},[36107],{"type":26,"value":36108},"\"zne\"",{"type":21,"tag":322,"props":36110,"children":36111},{"style":329},[36112],{"type":26,"value":367},{"type":21,"tag":322,"props":36114,"children":36115},{"style":3703},[36116],{"type":26,"value":4560},{"type":21,"tag":322,"props":36118,"children":36119},{"style":3638},[36120],{"type":26,"value":3678},{"type":21,"tag":322,"props":36122,"children":36123},{"style":348},[36124],{"type":26,"value":36125},"2048",{"type":21,"tag":322,"props":36127,"children":36128},{"style":329},[36129],{"type":26,"value":3738},{"type":21,"tag":22,"props":36131,"children":36132},{},[36133,36135,36140,36141,36146],{"type":26,"value":36134},"量子化学シミュレーションの詳細については、完全な ",{"type":21,"tag":29,"props":36136,"children":36137},{"href":14507},[36138],{"type":26,"value":36139},"PennyLane guide",{"type":26,"value":20934},{"type":21,"tag":29,"props":36142,"children":36143},{"href":19441},[36144],{"type":26,"value":36145},"HLQuantum algorithms reference",{"type":26,"value":36147}," をご覧ください。",{"type":21,"tag":703,"props":36149,"children":36150},{},[36151],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":36153},[36154,36155,36156,36157,36158,36159,36160],{"id":34077,"depth":335,"text":34080},{"id":34116,"depth":335,"text":34116},{"id":34489,"depth":335,"text":34489},{"id":34834,"depth":335,"text":34837},{"id":35338,"depth":335,"text":35338},{"id":35632,"depth":335,"text":35635},{"id":35991,"depth":335,"text":35991},"content:ja:blog:vqe-pennylane-practical-guide.md","ja\u002Fblog\u002Fvqe-pennylane-practical-guide.md","ja\u002Fblog\u002Fvqe-pennylane-practical-guide",{"_path":36165,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":36166,"description":36167,"date":36168,"author":11,"tags":36169,"readingTime":453,"body":36170,"_type":715,"_id":36625,"_source":717,"_file":36626,"_stem":36627,"_extension":720},"\u002Fja\u002Fblog\u002Funderstanding-quantum-error-correction","数式なしで理解する量子誤り訂正","なぜ量子コンピュータは誤りを起こすのか、誤り訂正はどのように機能するのか、そしてそれがフォールトトレラント量子コンピューティングの未来にとって何を意味するのか。","2026-01-15",[729,3302,266],{"type":18,"children":36171,"toc":36616},[36172,36177,36182,36187,36206,36211,36216,36249,36254,36260,36279,36298,36306,36318,36324,36334,36342,36347,36352,36358,36363,36381,36386,36391,36397,36402,36435,36440,36554,36559,36564,36574,36584,36594,36612],{"type":21,"tag":22,"props":36173,"children":36174},{},[36175],{"type":26,"value":36176},"量子コンピュータは誤りを起こします — それも大量に。典型的な超伝導量子ビットのゲートは0.1〜1%の誤り率を持ちます。1000ゲートの回路を実行すれば、間違った答えを得ることは事実上確実です。これは現在のハードウェアのバグではなく、根本的な課題です。量子誤り訂正（QEC）はその解決策であり、それを理解することは量子コンピューティングがどこへ向かっているのかを理解する鍵となります。",{"type":21,"tag":59,"props":36178,"children":36180},{"id":36179},"なぜ量子誤りは異なるのか",[36181],{"type":26,"value":36179},{"type":21,"tag":22,"props":36183,"children":36184},{},[36185],{"type":26,"value":36186},"古典コンピュータにも誤りはありますが、対処は簡単です。ビットを3回コピーして多数決を取ればよいのです。1つのコピーが反転しても、残りの2つが多数決で勝ちます。",{"type":21,"tag":22,"props":36188,"children":36189},{},[36190,36192,36197,36199,36204],{"type":26,"value":36191},"量子ビットはコピーできません — ",{"type":21,"tag":71,"props":36193,"children":36194},{},[36195],{"type":26,"value":36196},"複製不可能定理（no-cloning theorem）",{"type":26,"value":36198}," がそれを禁じています。重ね合わせを崩さずに量子ビットを測定して誤りをチェックすることはできません。しかも量子誤りは単なるビット反転だけではありません。",{"type":21,"tag":71,"props":36200,"children":36201},{},[36202],{"type":26,"value":36203},"位相誤り",{"type":26,"value":36205},"（|0⟩ と |1⟩ の間の相対位相がずれるもの）や、その両方の組み合わせも存在します。",{"type":21,"tag":59,"props":36207,"children":36209},{"id":36208},"量子誤りの3つの種類",[36210],{"type":26,"value":36208},{"type":21,"tag":22,"props":36212,"children":36213},{},[36214],{"type":26,"value":36215},"あらゆる量子誤りは、3つのパウリ演算子の組み合わせに分解できます。",{"type":21,"tag":3556,"props":36217,"children":36218},{},[36219,36229,36239],{"type":21,"tag":3560,"props":36220,"children":36221},{},[36222,36227],{"type":21,"tag":71,"props":36223,"children":36224},{},[36225],{"type":26,"value":36226},"X誤り（ビット反転）:",{"type":26,"value":36228}," |0⟩ → |1⟩ および |1⟩ → |0⟩。古典的なビット反転に相当します。",{"type":21,"tag":3560,"props":36230,"children":36231},{},[36232,36237],{"type":21,"tag":71,"props":36233,"children":36234},{},[36235],{"type":26,"value":36236},"Z誤り（位相反転）:",{"type":26,"value":36238}," |0⟩ → |0⟩、|1⟩ → −|1⟩。古典的な対応物はなく、位相に影響します。",{"type":21,"tag":3560,"props":36240,"children":36241},{},[36242,36247],{"type":21,"tag":71,"props":36243,"children":36244},{},[36245],{"type":26,"value":36246},"Y誤り:",{"type":26,"value":36248}," X誤りとZ誤りが同時に組み合わさったもの。",{"type":21,"tag":22,"props":36250,"children":36251},{},[36252],{"type":26,"value":36253},"量子ビットに対する任意の誤りは、X、Y、Z（および恒等演算子）の線形結合として表すことができます。これは、X誤りとZ誤りを独立に訂正できれば、あらゆる誤りを訂正できることを意味します — これはQECにおける重要な洞察です。",{"type":21,"tag":59,"props":36255,"children":36257},{"id":36256},"大きなアイデアコピーなしの冗長性",[36258],{"type":26,"value":36259},"大きなアイデア：コピーなしの冗長性",{"type":21,"tag":22,"props":36261,"children":36262},{},[36263,36265,36270,36272,36277],{"type":26,"value":36264},"QECは1つの",{"type":21,"tag":71,"props":36266,"children":36267},{},[36268],{"type":26,"value":36269},"論理量子ビット",{"type":26,"value":36271},"を多数の",{"type":21,"tag":71,"props":36273,"children":36274},{},[36275],{"type":26,"value":36276},"物理量子ビット",{"type":26,"value":36278},"にわたって符号化します。この考え方は古典的な繰り返し符号に似ていますが、量子力学に合わせて適応されています。",{"type":21,"tag":22,"props":36280,"children":36281},{},[36282,36284,36289,36291,36296],{"type":26,"value":36283},"最も単純な例は",{"type":21,"tag":71,"props":36285,"children":36286},{},[36287],{"type":26,"value":36288},"3量子ビットビット反転符号",{"type":26,"value":36290},"です。|0⟩ を |000⟩ として、|1⟩ を |111⟩ として符号化します。1つの量子ビットが反転した場合、（実際の値を測定せずに）量子ビットのペア間の",{"type":21,"tag":51,"props":36292,"children":36293},{},[36294],{"type":26,"value":36295},"パリティ",{"type":26,"value":36297},"を測定することで検出し、訂正します。",{"type":21,"tag":312,"props":36299,"children":36301},{"code":36300},"Logical |0⟩ = |000⟩\nLogical |1⟩ = |111⟩\n\nIf qubit 1 flips: |100⟩ or |011⟩\nParity check: measure (q0 ⊕ q1) and (q1 ⊕ q2)\n  → 10 → qubit 0 flipped → apply X to correct\n",[36302],{"type":21,"tag":79,"props":36303,"children":36304},{"__ignoreMap":7},[36305],{"type":26,"value":36300},{"type":21,"tag":22,"props":36307,"children":36308},{},[36309,36311,36316],{"type":26,"value":36310},"パリティ測定には",{"type":21,"tag":71,"props":36312,"children":36313},{},[36314],{"type":26,"value":36315},"補助量子ビット（ancilla qubits）",{"type":26,"value":36317}," を使用します — これはデータ量子ビットを乱すことなく誤り情報を吸収する補助的な量子ビットです。これこそが魔法です。符号化された状態を崩すことなく誤りを検出できるのです。",{"type":21,"tag":59,"props":36319,"children":36321},{"id":36320},"表面符号今日の主要なアプローチ",[36322],{"type":26,"value":36323},"表面符号：今日の主要なアプローチ",{"type":21,"tag":22,"props":36325,"children":36326},{},[36327,36332],{"type":21,"tag":71,"props":36328,"children":36329},{},[36330],{"type":26,"value":36331},"表面符号（surface code）",{"type":26,"value":36333}," は、近い将来のハードウェアにとって最も有望なQEC符号です。これは1つの論理量子ビットを、次のように配置された物理量子ビットの2Dグリッドに符号化します。",{"type":21,"tag":312,"props":36335,"children":36337},{"code":36336},"● — ● — ● — ●\n|   |   |   |\n● — ● — ● — ●\n|   |   |   |\n● — ● — ● — ●\n|   |   |   |\n● — ● — ● — ●\n",[36338],{"type":21,"tag":79,"props":36339,"children":36340},{"__ignoreMap":7},[36341],{"type":26,"value":36336},{"type":21,"tag":22,"props":36343,"children":36344},{},[36345],{"type":26,"value":36346},"データ量子ビット（●）が量子情報を保持します。その間にある補助量子ビットが継続的にパリティを測定します。7×7のグリッド（49個の物理量子ビット）は、誤り率 ~p² で1つの論理量子ビットを符号化します — 物理ゲート誤り率が1%であれば、論理誤り率は ~0.01% まで下がります。",{"type":21,"tag":22,"props":36348,"children":36349},{},[36350],{"type":26,"value":36351},"表面符号の閾値は約1%です。ハードウェアのゲート誤りがこれを下回っていれば、物理量子ビットを増やすことで論理誤り率を任意に小さくできます。IBMの最良のシステムはこの閾値に近づきつつあります。",{"type":21,"tag":59,"props":36353,"children":36355},{"id":36354},"何個の物理量子ビットが必要か",[36356],{"type":26,"value":36357},"何個の物理量子ビットが必要か？",{"type":21,"tag":22,"props":36359,"children":36360},{},[36361],{"type":26,"value":36362},"RSA-2048を破るためにショアのアルゴリズムを誤りなく実行するには、次のように推定されています。",{"type":21,"tag":3556,"props":36364,"children":36365},{},[36366,36371,36376],{"type":21,"tag":3560,"props":36367,"children":36368},{},[36369],{"type":26,"value":36370},"約2000万個の物理量子ビット",{"type":21,"tag":3560,"props":36372,"children":36373},{},[36374],{"type":26,"value":36375},"各論理量子ビットには約1000〜10,000個の物理量子ビットが必要",{"type":21,"tag":3560,"props":36377,"children":36378},{},[36379],{"type":26,"value":36380},"ゲート誤り率は0.1%未満でなければならない",{"type":21,"tag":22,"props":36382,"children":36383},{},[36384],{"type":26,"value":36385},"現在のハードウェア：約1000量子ビット、約0.1〜1%のゲート誤り。ギャップは現実のものです — フォールトトレラントなRSA破りの量子コンピュータまでは、おおよそ10〜15年です。",{"type":21,"tag":22,"props":36387,"children":36388},{},[36389],{"type":26,"value":36390},"化学や最適化における近い将来の量子優位性については、要件ははるかに低くなります。約100個の論理量子ビットがあれば、古典的な計算の手に負えない分子に対して十分かもしれません。",{"type":21,"tag":59,"props":36392,"children":36394},{"id":36393},"これが今日のnisqデバイスにとって何を意味するのか",[36395],{"type":26,"value":36396},"これが今日のNISQデバイスにとって何を意味するのか",{"type":21,"tag":22,"props":36398,"children":36399},{},[36400],{"type":26,"value":36401},"現在のNISQデバイスにはQECがありません。すべてのゲートがノイズを加えます。代わりのアプローチは次のとおりです。",{"type":21,"tag":7444,"props":36403,"children":36404},{},[36405,36415,36425],{"type":21,"tag":3560,"props":36406,"children":36407},{},[36408,36413],{"type":21,"tag":71,"props":36409,"children":36410},{},[36411],{"type":26,"value":36412},"回路を浅く保つ",{"type":26,"value":36414}," — ゲートが少ないほど、蓄積されるノイズが少ない",{"type":21,"tag":3560,"props":36416,"children":36417},{},[36418,36423],{"type":21,"tag":71,"props":36419,"children":36420},{},[36421],{"type":26,"value":36422},"誤り緩和を使う",{"type":26,"value":36424},"（訂正ではなく） — 追加の量子ビットなしで統計的に誤りの影響を減らす、Zero Noise Extrapolation（ZNE）などの後処理技術",{"type":21,"tag":3560,"props":36426,"children":36427},{},[36428,36433],{"type":21,"tag":71,"props":36429,"children":36430},{},[36431],{"type":26,"value":36432},"ノイズ耐性のあるアルゴリズムを設計する",{"type":26,"value":36434}," — VQE、QAOA、その他の変分法は、中程度のノイズに対して頑健になるよう設計されている",{"type":21,"tag":22,"props":36436,"children":36437},{},[36438],{"type":26,"value":36439},"HLQuantumには、任意の回路に適用できる組み込みの誤り緩和機能が含まれています。",{"type":21,"tag":312,"props":36441,"children":36443},{"code":36442,"language":3628,"meta":7,"className":3626,"style":7},"result = hlq.run(\n    qc,\n    backend=\"qiskit\",\n    device=\"ibm_sherbrooke\",\n    error_mitigation=\"zne\",  # Zero Noise Extrapolation\n    shots=8192\n)\n",[36444],{"type":21,"tag":79,"props":36445,"children":36446},{"__ignoreMap":7},[36447,36462,36469,36488,36507,36531,36547],{"type":21,"tag":322,"props":36448,"children":36449},{"class":324,"line":325},[36450,36454,36458],{"type":21,"tag":322,"props":36451,"children":36452},{"style":329},[36453],{"type":26,"value":19764},{"type":21,"tag":322,"props":36455,"children":36456},{"style":3638},[36457],{"type":26,"value":3678},{"type":21,"tag":322,"props":36459,"children":36460},{"style":329},[36461],{"type":26,"value":29182},{"type":21,"tag":322,"props":36463,"children":36464},{"class":324,"line":335},[36465],{"type":21,"tag":322,"props":36466,"children":36467},{"style":329},[36468],{"type":26,"value":29190},{"type":21,"tag":322,"props":36470,"children":36471},{"class":324,"line":344},[36472,36476,36480,36484],{"type":21,"tag":322,"props":36473,"children":36474},{"style":3703},[36475],{"type":26,"value":23300},{"type":21,"tag":322,"props":36477,"children":36478},{"style":3638},[36479],{"type":26,"value":3678},{"type":21,"tag":322,"props":36481,"children":36482},{"style":359},[36483],{"type":26,"value":29206},{"type":21,"tag":322,"props":36485,"children":36486},{"style":329},[36487],{"type":26,"value":428},{"type":21,"tag":322,"props":36489,"children":36490},{"class":324,"line":407},[36491,36495,36499,36503],{"type":21,"tag":322,"props":36492,"children":36493},{"style":3703},[36494],{"type":26,"value":29218},{"type":21,"tag":322,"props":36496,"children":36497},{"style":3638},[36498],{"type":26,"value":3678},{"type":21,"tag":322,"props":36500,"children":36501},{"style":359},[36502],{"type":26,"value":29227},{"type":21,"tag":322,"props":36504,"children":36505},{"style":329},[36506],{"type":26,"value":428},{"type":21,"tag":322,"props":36508,"children":36509},{"class":324,"line":431},[36510,36514,36518,36522,36526],{"type":21,"tag":322,"props":36511,"children":36512},{"style":3703},[36513],{"type":26,"value":29265},{"type":21,"tag":322,"props":36515,"children":36516},{"style":3638},[36517],{"type":26,"value":3678},{"type":21,"tag":322,"props":36519,"children":36520},{"style":359},[36521],{"type":26,"value":36108},{"type":21,"tag":322,"props":36523,"children":36524},{"style":329},[36525],{"type":26,"value":29232},{"type":21,"tag":322,"props":36527,"children":36528},{"style":3888},[36529],{"type":26,"value":36530},"# Zero Noise Extrapolation\n",{"type":21,"tag":322,"props":36532,"children":36533},{"class":324,"line":453},[36534,36538,36542],{"type":21,"tag":322,"props":36535,"children":36536},{"style":3703},[36537],{"type":26,"value":29245},{"type":21,"tag":322,"props":36539,"children":36540},{"style":3638},[36541],{"type":26,"value":3678},{"type":21,"tag":322,"props":36543,"children":36544},{"style":348},[36545],{"type":26,"value":36546},"8192\n",{"type":21,"tag":322,"props":36548,"children":36549},{"class":324,"line":474},[36550],{"type":21,"tag":322,"props":36551,"children":36552},{"style":329},[36553],{"type":26,"value":3738},{"type":21,"tag":22,"props":36555,"children":36556},{},[36557],{"type":26,"value":36558},"ZNEは複数のレベルで意図的にノイズを増幅し、ゼロノイズ極限へと外挿します。これはQECではなく — 統計的なトリックです — しかし短い回路に対しては驚くほどよく機能します。",{"type":21,"tag":59,"props":36560,"children":36562},{"id":36561},"注目すべき近い将来のマイルストーン",[36563],{"type":26,"value":36561},{"type":21,"tag":22,"props":36565,"children":36566},{},[36567,36572],{"type":21,"tag":71,"props":36568,"children":36569},{},[36570],{"type":26,"value":36571},"スケールでの誤り訂正の実証",{"type":26,"value":36573}," — IBM、Google、Microsoftは、表面符号の論理量子ビットが物理量子ビットを上回る（誤り率が閾値を下回る）ことを実証しようと競っています。Googleは2024年後半に閾値を下回る実証を主張しました。",{"type":21,"tag":22,"props":36575,"children":36576},{},[36577,36582],{"type":21,"tag":71,"props":36578,"children":36579},{},[36580],{"type":26,"value":36581},"マジック状態蒸留",{"type":26,"value":36583}," — 非クリフォードゲート（Tゲートなど）は、フォールトトレラントなアーキテクチャにおいて特別な蒸留プロトコルを必要とします。これは現在、量子ビットのオーバーヘッドの面で極めて高コストです。",{"type":21,"tag":22,"props":36585,"children":36586},{},[36587,36592],{"type":21,"tag":71,"props":36588,"children":36589},{},[36590],{"type":26,"value":36591},"論理ゲートの速度",{"type":26,"value":36593}," — 表面符号における論理ゲートは、物理ゲートよりもはるかに遅いです。それを高速化することは、2026〜2030年における重要なハードウェアの課題です。",{"type":21,"tag":22,"props":36595,"children":36596},{},[36597,36599,36604,36605,36610],{"type":26,"value":36598},"誤り訂正が特定のSDKとどのように相互作用するかについてのより深い解説は、",{"type":21,"tag":29,"props":36600,"children":36601},{"href":1526},[36602],{"type":26,"value":36603},"量子誤り訂正の用語集エントリ",{"type":26,"value":811},{"type":21,"tag":29,"props":36606,"children":36607},{"href":263},[36608],{"type":26,"value":36609},"NISQ時代ガイド",{"type":26,"value":36611},"を参照してください。",{"type":21,"tag":703,"props":36613,"children":36614},{},[36615],{"type":26,"value":707},{"title":7,"searchDepth":335,"depth":335,"links":36617},[36618,36619,36620,36621,36622,36623,36624],{"id":36179,"depth":335,"text":36179},{"id":36208,"depth":335,"text":36208},{"id":36256,"depth":335,"text":36259},{"id":36320,"depth":335,"text":36323},{"id":36354,"depth":335,"text":36357},{"id":36393,"depth":335,"text":36396},{"id":36561,"depth":335,"text":36561},"content:ja:blog:understanding-quantum-error-correction.md","ja\u002Fblog\u002Funderstanding-quantum-error-correction.md","ja\u002Fblog\u002Funderstanding-quantum-error-correction"]