{"version":"1.0","publisher":"qubit-lab.ch","page":"https://qubit-lab.ch/evidence","license":"Figures are unedited solver output on demo data — illustrative only, not investment advice.","verdictVocabulary":["Advantage demonstrated","Parity","No advantage demonstrated","Inconclusive","Method study"],"records":[{"id":"QL-EV-001","product":"Portfolio Optimization RQP (QAOA)","reportDate":"16 Jul 2026","verdict":"Parity","fields":[{"k":"Problem","v":"Long-only portfolio selection under 3 sector-budget bands"},{"k":"Instance","v":"78 assets → 23 binary decisions (17 single names + 6 thematic bundles); 2²³ = 8,388,608 candidate portfolios"},{"k":"Data","v":"78-asset demo universe — illustrative, not investment advice"},{"k":"Classical referee","v":"Exact branch & bound (MIQP, SCIP) — certified optimum over the full space"},{"k":"Referee result","v":"Expected return 8.0% · volatility 15.3% · Sharpe 0.33 · budget gap 0.08%"},{"k":"Classical baseline","v":"Heuristic search, 465 candidates — +3.2% gap, optimum not reached"},{"k":"Quantum method","v":"QAOA, depth p=4, warm-started, seed 42"},{"k":"Execution","v":"Statevector simulator, 23 qubits, exact readout (no sampling) · 2 h 28 min certified end-to-end"},{"k":"Quantum result","v":"Gap 0.0% — best QAOA portfolio equals the SCIP-proven optimum, bit for bit · P(optimum) 0.0016% = 137× uniform"},{"k":"Verdict basis","v":"Solution quality matched the certified optimum at a scale where classical enumeration is exact; advantage not claimed"},{"k":"Limitations","v":"Optimum sampling probability 0.0016% — a boost, still small · simulator memory doubles per qubit, ~30 is the hardware line · demo data"},{"k":"Reproducibility","v":"Seed 42 · certified PDF report in the RQP Resource Room"}]},{"id":"QL-EV-002","product":"Index Replication RQP (QAOA)","reportDate":"12 Jul 2026","verdict":"Parity","fields":[{"k":"Problem","v":"Enhanced indexing of the S&P 500 with 20 instruments under mandate rules (technology band, climate-transition band)"},{"k":"Instance","v":"500 assets in the objective, 17 selection qubits, 20-name cardinality; demo mandate, 20 ISINs"},{"k":"Data","v":"S&P 500 demo mandate — illustrative, not investment advice"},{"k":"Classical referee","v":"SCIP dual bound, MIP gap 0 — certified minimum tracking error; each rule priced by re-solving with that rule removed"},{"k":"Referee result","v":"Certified minimum tracking error 4.473% · irreducible floor 3.098% from pool and exclusions alone"},{"k":"Classical baseline","v":"Greedy heuristic (industry-standard shortcut) +3.39% · random baseline +1.20%"},{"k":"Quantum method","v":"QAOA, depth p=2, 3 restarts, exact readout"},{"k":"Execution","v":"17 qubits, 544 two-qubit gates · 110 s end-to-end (QAOA 53.8 s, referee proof 11.2 s)"},{"k":"Quantum result","v":"Gap 0.00% — its basket is a certified-optimal one, after repair and a single disclosed polish swap"},{"k":"Verdict basis","v":"Matched the certified optimum; 3.6% of raw readout met the cardinality — repair pipeline disclosed; advantage not claimed"},{"k":"Limitations","v":"On today's hardware this circuit would retain ~7% fidelity · candidates repaired before scoring · demo data"},{"k":"Reproducibility","v":"Certified PDF report in the RQP Resource Room"}]},{"id":"QL-EV-003","product":"Robust Stochastic Collateral Allocation RQP (CVaR QAOA)","reportDate":"10 Aug 2026","verdict":"Parity","fields":[{"k":"Problem","v":"Collateral allocation across three venues under margin calls — posting cost vs coverage stability via a scenario-covariance term"},{"k":"Instance","v":"23 collateral lines (10 locked, 3 pledge pools moving whole) · 50 stress scenarios · 24-variable register; 2²⁴ = 16,777,216 allocations"},{"k":"Data","v":"Swiss bank treasury demo workbook — illustrative"},{"k":"Classical referee","v":"Exhaustive enumeration of every valid allocation — proven optimum"},{"k":"Referee result","v":"Certified optimum: govt-bond ladder to LCH · Pfandbrief pool to Eurex · equity basket to the CSA · 14% scenario undercoverage probability, printed"},{"k":"Classical baseline","v":"Best classically-enumerated allocation used as warm-start incumbent (stated in report)"},{"k":"Quantum method","v":"Warm-started CVaR QAOA, depth p=3, fixed seeds"},{"k":"Execution","v":"Exact statevector readout of all 16.8M states · 29 min end-to-end"},{"k":"Quantum result","v":"Gap 0.00% — the optimum became the modal state at 2.6% (~428,000× uniform) · frontier flip priced at CHF 19k/yr (flagship run)"},{"k":"Verdict basis","v":"Concentrated on the enumeration-proven optimum from a stated warm start; advantage not claimed"},{"k":"Limitations","v":"Stress panels are classical scenario analysis, labeled · 36-variable flagship exceeds the simulator — quantum stage skipped by design · early failures (infeasible best allocation, too-coarse slack encoding) published"},{"k":"Reproducibility","v":"Fixed seeds · certified PDF report of 10 Aug 2026 in the RQP Resource Room"}]},{"id":"QL-EV-004","product":"QML Classification RQP (quantum kernel)","reportDate":"23 Jul 2026 · re-verified 26 Jul 2026","verdict":"No advantage demonstrated","fields":[{"k":"Problem","v":"Classifying German power-price spikes"},{"k":"Instance","v":"17,157 rows · exact 10,294×10,294 quantum kernel · 8-qubit feature map · QML suitability score 34/100 (±12): “too simple for quantum”"},{"k":"Data","v":"German power-price spike demo dataset (public), baseline & Fourier-engineered features"},{"k":"Classical referee","v":"Six tuned classical baselines incl. an order-matched periodic twin to interaction order 3 (Fourier Wall methodology, arXiv:2607.15815) · paired bootstrap"},{"k":"Referee result","v":"Classical ahead on raw features; advantage claimed only if the whole 95% CI clears 1.0"},{"k":"Classical baseline","v":"Included in the referee bar (six tuned models + classical-stack ensemble control)"},{"k":"Quantum method","v":"Quantum kernel classifier, 8 qubits, exact kernel evaluation"},{"k":"Execution","v":"Two-level assessment gate: Fast (seconds) and Deep (minutes) before any training run"},{"k":"Quantum result","v":"Advantage factor 0.92× raw (95% CI 0.896–0.937, below 1.0) · 0.99× Fourier-engineered (95% CI 0.975–1.011, includes 1.0)"},{"k":"Verdict basis","v":"Printed verdict: keep the classical model for this dataset, today — inconclusive within noise after encoding gains; five-condition checklist: still classical territory"},{"k":"Limitations","v":"A verdict about one dataset, not about quantum · re-tested as encodings and hardware evolve"},{"k":"Reproducibility","v":"Certified PDF reports in the RQP Resource Room · upgraded classical bar re-verification 26 Jul 2026"}]},{"id":"QL-EV-005","product":"QML Classification RQP (quantum kernel + VQC)","reportDate":"26 Jul 2026","verdict":"Advantage demonstrated","fields":[{"k":"Problem","v":"Binary classification on a quantum-native demo dataset (quantum-teacher generated)"},{"k":"Instance","v":"Same assessment gate and classical bar as QL-EV-004, passed"},{"k":"Data","v":"Quantum-native demo dataset — full report and dataset in the Resource Room"},{"k":"Classical referee","v":"Full classical bar incl. order-matched twin (Fourier Wall methodology) · paired bootstrap 95% CI"},{"k":"Referee result","v":"Best classical F1 0.71"},{"k":"Classical baseline","v":"Included in the referee bar"},{"k":"Quantum method","v":"Quantum kernel + 6-qubit variational quantum classifier"},{"k":"Execution","v":"Same two-level gate as QL-EV-004"},{"k":"Quantum result","v":"Advantage factor 1.45× (95% CI 1.31–1.62, entirely above 1.0) · VQC test F1 0.89 vs 0.71 best classical"},{"k":"Verdict basis","v":"Whole 95% CI above 1.0 against the full classical bar, order-matched twin included"},{"k":"Limitations","v":"Quantum-native demo data — engineered to sit in quantum territory; scope is the gate working in both directions, not a market dataset win"},{"k":"Reproducibility","v":"Full report and dataset in the RQP Resource Room"}]},{"id":"QL-ST-001","product":"QAOA Concentration Study (method study)","reportDate":"15–16 Aug 2026","verdict":"Method study","fields":[{"k":"Question","v":"Which QAOA method concentrates probability on the certified optimum — and under what prior knowledge?"},{"k":"Design","v":"Training objective (expected energy vs CVaR) × ansatz (fixed depth vs adaptive) × start state (uniform vs classical seed) × instance · matched evaluation budgets · 3 optimizer seeds per cell · pre-registered with amendments recorded"},{"k":"Instances","v":"24 instances: 20 synthetic dense Markowitz at 12 qubits + 4 real demo books at 7–16 qubits · 12,397 certified runs"},{"k":"Classical referee","v":"Exhaustive enumeration — every run scored against the proven optimum"},{"k":"Key results","v":"Penalty ratio vs concentration: Spearman −0.91 (CVaR; 95% CI −0.95 to −0.80) · classical seed: right seed 100% of runs, wrong seed 0 of 96 (below 3.1% at 95% confidence) · adaptive ansatz escapes a wrong seed 83/75/50% (1/2/3 assets off, real books, n = 12 per distance — Wilson intervals printed in the study)"},{"k":"Null results (printed)","v":"Matched warm-start mixer vs plain X mixer: almost no change — the lock lives in the starting state · adaptive ansatz never wins on efficiency at matched budgets (~290 evaluations per layer)"},{"k":"Scope","v":"Exact statevector simulation only — no hardware: shot noise would make the adaptive method's screening worse, not better"},{"k":"Reproducibility","v":"Study PDF (18 pages, canonical, Rev. 1.2 — uncertainty intervals on every headline statistic) linked on this page · headline numbers reproduced from committed raw results by a committed script · raw results and design on request"}]},{"id":"QL-ST-002","product":"QAOA Noise-Ladder Study (method study)","reportDate":"21–22 Aug 2026","verdict":"Method study","fields":[{"k":"Question","v":"How much of a trained circuit's concentration survives noise as depth grows — and does a calibrated simulation predict the real device?"},{"k":"Design","v":"5 trained circuits (standard p = 1, 3, 5, 7 + adaptive cap 12; CVaR, classical seed start) replayed unchanged at 3 parametric noise levels (2-qubit fidelity 99.9 / 99.5 / 99.0%), a calibrated simulation of ibm_kingston and ibm_kingston itself · pre-registered, 6 amendments recorded · 100-hit statistical floor"},{"k":"Instance","v":"Real 14-qubit demo book (16,384 states); classical seed wrong by 3 assets (certified gap 1.60%)"},{"k":"Classical referee","v":"SCIP branch-and-bound, certified optimum proven on every run — every probability measured against it"},{"k":"Hardware","v":"ibm_kingston (IBM Heron r2, 156 qubits), 20,000 shots per circuit, raw (no error mitigation), 44 s of processor time in total · routing onto the heavy-hex lattice multiplied two-qubit gates by ~2.5 (182 → 440 … 2,190 → 5,454)"},{"k":"Key results","v":"Exact P(optimum) saturates at p ≈ 5 (0.065% → 0.170%) · seed-locked circuits smear before they erase: retention 1.0–1.5 at moderate noise while top-10 mass falls monotonically · adaptive circuit (10.08% exact, modal) keeps the optimum modal at 99.9% (2.15%) and is depolarised at ≤ 99.5% · device: only p = 1 retains structure (seed peak 8.5% → 0.8%), p ≥ 3 statistically uniform, adaptive 0 hits in 20,000 (below 0.015% at 95% confidence) · calibration-based prediction optimistic ×2.6–3.5 (shot-noise intervals stay above ×2.3)"},{"k":"Null results (printed)","v":"Prediction of geometric decay per layer was wrong in the smear regime · no configuration delivers a usable readout below 99.9% two-qubit fidelity after routing on this book"},{"k":"Cost (printed)","v":"Device runs: 44 s of processor time for 100,000 shots, queue waits 5–10 s · calibrated prediction: 17–26 min of trajectory sampling per circuit on the standard worker (gate + readout errors; thermal relaxation excluded by design) · one prediction run repeated after a transient IBM account error"},{"k":"Reproducibility","v":"Study PDF (16 pages, canonical, Rev. 1.3 — finite-shot intervals on every headline statistic) linked on this page · IBM job ids printed in the study · headline numbers reproduced from committed raw counts by a committed script · raw results, design and scripts on request"}]},{"id":"QL-ST-003","product":"QML Advantage-Island Study (method study)","reportDate":"27–28 Aug 2026","verdict":"Method study","fields":[{"k":"Question","v":"Where can a quantum kernel classifier (QSVM) beat a full classical referee at all — between the classical-learnability wall and the kernel-concentration wall?"},{"k":"Design","v":"996 pre-specified referee arms: label structure (prototype density + a mixing dial toward twin-representable classical structure) × register size (4–16 qubits) × kernel estimation (exact / 4,096 / 512 shots) · 6 replicates per cell · design locked in writing before the grids, amendments and deviations disclosed"},{"k":"Labels (disclosed)","v":"Quantum-native by construction — generated by the same fidelity kernel the classifier uses: a best-case benchmark envelope for this feature-map family, not a real-world claim"},{"k":"Classical referee","v":"Ten variants, each threshold-tuned: logistic regression, random forest, gradient boosting, order-matched periodic twin (parity + best-reference), tuned RBF kernel SVM, neural network · paired-bootstrap Quantum Advantage Factor, win only when the whole 95% CI clears 1.0 · selection-aware panel-max bootstrap confirms 98% of wins"},{"k":"Key results","v":"Wins in ≥94% of arms at every size up to 12 qubits — at 512 shots as at exact simulation; erosion at 14–16 qubits · the per-entry 3/√shots heuristic predicts failure ~2 register sizes too early (binomial small-K variance, learner aggregation, bulk/tail redundancy — measured by ablation) · classical wall located: 0 wins on pure low-order labels, wins begin at 75% quantum-native mixing · expanded referee flipped zero verdicts"},{"k":"Null results (printed)","v":"Lower-wall hypothesis refuted (even one prototype beats the twin) · strong tail-mechanism conjecture refuted by its own ablation (exact kernels are redundant across bulk and tail) · own screening checklist vetoed 83 of 84 winning datasets — replaced by calibrated measurements (kernel statistics rank outcomes at AUC up to 0.97)"},{"k":"Scope","v":"Kernel methods only (variational classifier: dedicated follow-up study) · exact or binomially shot-sampled simulation, no hardware runs · one feature-map family (zz-like, 2 repeats)"},{"k":"Reproducibility","v":"Study PDF (14 pages, canonical) linked on this page · every number re-derived from committed raw results by committed scripts · pipeline deterministic — the production demo run reproduces study values to four decimals"}]},{"id":"QL-ST-005","product":"QML Kernel Configuration Atlas Study (method study)","reportDate":"6–7 Sep 2026","verdict":"Method study","fields":[{"k":"Question","v":"Which settings of the quantum kernel classifier (QSVM) can be measured at all, which are dead on arrival — and does measurable mean winning?"},{"k":"Design","v":"5 feature maps (angle-X/Y/Z, iqp, zz_like) × 4 repeats × 5 encoding bandwidths × 4 register sizes (4–16 qubits) × 3 label families × 6 replicate datasets = 7,200 kernel-concentration measurements with the product's own diagnostic, plus 150 refereed configurations drawn by a pre-registered stratified sampler · design tagged (kernel-config-atlas-design-v1) AND deposited with Zenodo (DOI 10.5281/zenodo.22498143) before the first run — the series' first deposit-first study"},{"k":"Labels (disclosed)","v":"The island studies' synthetic families, unchanged: best-case envelopes, not real-world data — the atlas maps configurations, not datasets"},{"k":"Referee","v":"Ten classical variants, threshold-tuned, whole-95%-CI win rule, selection-aware panel-max — and, post-hoc, the exact closed-form classical twin of the angle map (identity verified to 1e-15)"},{"k":"Key results","v":"Measurable at 512 shots: 100% / 86% / 33% / 25% of configurations at 4 / 8 / 12 / 16 qubits (36% at 32,768 shots at 16q) · repeats and entangling maps concentrate as predicted (H1 supported, 48/48 groups) · the angle map with rotation Z encodes nothing: kernel identically 1 in all 1,440 cells, read as 'resolvable' by the diagnostic · of 150 refereed configurations 4 met the win rule at exact simulation — all angle maps, 0 once their exact classical twin joins the referee · off-native configurations memorize (train 0.99, test near chance)"},{"k":"Null results (printed)","v":"Registered bet H2 (bandwidth beats structure) refuted on the pre-registered grid (map factor dominates); post-hoc sensitivity excluding the dead map disclosed · H3 inconclusive (one shot-noise flip) · H4 refuted (0/4 winners near the shot-budget line) · H5 refuted for Z (degenerate), exact for X vs Y · the sampler did not land on study 3's known winning configuration"},{"k":"Scope","v":"Exact and shot-sampled simulation, no hardware · one kernel family (fidelity) · 4–16 qubits · threshold tuning collapses weak models to one class (101/150) — verdict directions robust, absolute scores deflated, disclosed"},{"k":"Reproducibility","v":"Study PDF (10 pages, canonical) linked on this page · every number composed from committed tables by a committed script · raw atlas (7,200 rows) and validation results committed · four product changes shipped from the findings (angle-Z rejected, degenerate-kernel flag, separable twin in the referee, bandwidth knob bounded)"}]},{"id":"QL-ST-004","product":"QML VQC Island Study (method study)","reportDate":"29–30 Aug 2026","verdict":"Method study","fields":[{"k":"Question","v":"Does the trainable quantum model (VQC) own territory the quantum kernel method does not — when is it reasonable, what does a fair attempt cost, and what can be screened before training?"},{"k":"Design","v":"65 pre-specified fair-recipe training arms (registered ceiling 150; 24 training-hours, disclosed) across three label families (VQC-native teacher, kernel-native, classical mixture) × 4–16 qubits, plus registered knob probes and instrument calibration · design publicly pushed and tagged BEFORE the first run (git tag vqc-island-design-v1)"},{"k":"Labels (disclosed)","v":"Native families are best-case envelopes per method; the teacher generator is deliberately UNSCREENED against the classical bar (unlike the certified sample) to avoid biasing the territory test"},{"k":"Referee & head-to-head","v":"Same ten classical variants as the kernel study, threshold-tuned, whole-95%-CI verdict rule with selection-aware robustness · plus the QSVM run on identical rows in every cell — the paired VQC/QSVM ratio is the territory readout"},{"k":"Key results","v":"Territory is size-gated with a monotone crossover: paired VQC/QSVM 0.98 → 1.18 across 4–16 qubits on VQC-native labels (QSVM perfect at 4 qubits, VQC ahead from 12); kernel-native territory stays the QSVM's · no trainability wall observed through 16 qubits (init gradients flat 0.055–0.092; best native score AT 16 qubits) · 512-shot inference noise negligible (median ΔF1 −0.004 with the exact-trained threshold held fixed) · a fair run = full depth + proportionate stopping — 150 scaled iterations already hold verdicts on matched structure"},{"k":"Null results (printed)","v":"Registered 14–16-qubit degradation prediction refuted on native territory · registered anti-folklore bet lost: initial-gradient variance predicts outcomes (AUC 0.79, interval above chance) instead of failing · head refitting mildly hurts · verdict-flip-by-budget prediction refuted · knobs cannot buy territory (every variant on mismatched structure stays classical-ahead)"},{"k":"Scope","v":"Exact-statevector training (training under shots: registered future work) · one shallow local-cost ansatz family — the regime the barren-plateau fine print permits; no claim beyond it · small-N instrument calibration (16 datasets), intervals printed"},{"k":"Reproducibility","v":"Study PDF (10 pages, canonical) linked on this page · design tag provably precedes every run · raw per-arm results incl. wall-clock and persisted model parameters committed"}]}]}