Complete documentation catalog

July 29, 2026 · View on GitHub

The primary navigation stays intentionally compact. This catalog keeps every public guide, evidence note, contract, campaign protocol, and reference page discoverable without crowding the main learning path.

Current inventory: 366 public pages; 162 are in the primary navigation and the remainder are indexed here.

Labels marked catalog are intentionally outside the primary navigation; they are still built, link-checked, searchable, and public.

API and contracts

PageSurfacePurpose
Analysis API Referenceprimary navAdvanced references for the analysis stack. The governed research-lane registry currently classifies 74 ordinary analysis/ and gauge/ modules by maturity, differentiable relevance, and claim status.
API Referenceprimary navThis API page is divided by abstraction level. Use it to decide whether you need a stable entry point for integration, a workflow guide for experimentation, or a source-level reference for extension and debugging.
Auto-Generated API Referenceprimary navGenerated from source docstrings via mkdocstrings.
Benchmarks API Referenceprimary navThe benchmarks package measures the computational frontier: at what system size does quantum hardware outperform the best classical methods for simulating Kuramoto-XY dynamics? Seven modules answer this from different angles — documented cl
Bridges API Referenceprimary navThe bridge package is the central nervous system of scpn-quantum-control. Every module in this package translates between classical SCPN state representations and quantum operator formats. Without the bridge layer, coupling matrices are num
Cross-language API reference buildsprimary navThe repository builds API references with each language ecosystem's native documentation tool. CI treats warnings and broken links as failures. Generated HTML stays out of Git; source comments, docstrings, and configuration remain the revie
Differentiable APIprimary nav> New here? Start with Choosing a Gradient Path — > a 60-second decision table from what you are differentiating to which entry > point to call. This page is the full reference behind that guide.
Identity Continuity APIprimary navThe identity package provides quantitative tools for characterising identity attractor basins, coherence budgets, entanglement structure, robustness certification, and cryptographic fingerprinting of coupling topologies.
Mitigation API Referenceprimary navThe mitigation package provides six error mitigation techniques for noise reduction on NISQ hardware. Each technique exploits a different aspect of the noise: ZNE extrapolates through noise amplification, PEC cancels errors via quasi-probab
Phase Evolution API Referenceprimary nav14 modules implementing variational, adiabatic, Floquet, and time-evolution algorithms for the Kuramoto-XY Hamiltonian on quantum hardware.
Public API stability programme (BL-97 / W5)catalogFail-closed public-vs-internal stability catalogue for adopters. Declares a narrow durable SemVer-intent surface (stable_core + curated CLI entry points) and classifies workbench / private paths so bulk package exports are not s
Stable Core APIprimary navThe stable core API defines the first durable contracts for production workflows:
Stable Facades APIprimary navStable facades are the first-path API surfaces for users who want to build a workflow without depending on low-level module layout. Prefer these symbols in tutorials, notebooks, and inter-repository contracts.

Api

PageSurfacePurpose
Application Honesty APIprimary navscpn_quantum_control.applications.honesty_kits is the public BL-63 policy surface for immutable domain claim boundaries and deterministic evidence. scpn_quantum_control.applications.dataset_catalog owns the complementary packaged-artifact p
Chimera Control APIprimary navThe scpn_quantum_control.chimera_control package is the public BL-60 facade for finite synthetic chimera generation, nested coherence observables, analytic hierarchy targets, unapplied phase proposals, topology-constraint composition, and d
DLA Topology Control APIprimary navThe scpn_quantum_control.dla_topology_control package is the public BL-54 facade for finite parity-sector derivatives, fixed-active-set topology-ledger sensitivities, a synthetic projected-gradient task, and deterministic evidence custody.
Complete Python module and API catalogprimary navThis static catalog makes every ordinary Python module and every public module-level class or function discoverable without importing optional provider, accelerator, or scientific dependencies.
Research-lane registry APIprimary navModule: scpn_quantum_control.analysis.research_lane_registry
RL research governance APIprimary navModule: scpn_quantum_control.analysis.rl_research_governance
Theory-Hook Promotion APIprimary navImport the public surface from scpn_quantum_control.analysis:
Topology Kernel Product APIprimary navscpn_quantum_control.topology_kernel_product is the public BL-88 facade for bounded topology-aware fidelity kernels, custody-checked kernel ridge classification, deterministic graph controls, and frozen synthetic evidence.

Campaigns

PageSurfacePurpose
Campaign Artefactsprimary navThis directory contains dated experiment planning, preregistration, readiness, manifest, and result notes for hardware and simulation campaigns.
Adaptive FIM QPU Protocol BoundarycatalogDate: 2026-05-06
Mitigated Bell Re-Run Preregistration (KIMI-9)catalogDate: 2026-07-16
S5 Phase 1 Benchmark HarnesscatalogThis artefact records a no-QPU open-data reproduction of the Phase 1 DLA-parity dataset.
Benchmark Harness RegistrycatalogThis registry distinguishes implemented public benchmark harnesses from planned entries. Planned rows are visible so the roadmap is transparent, but they are not treated as available benchmark results.
Depth-Optimal Native Decomposition PreregistrationcatalogDate: 2026-05-06
DLA Multi-Device Replication PreregistrationcatalogDate: 2026-05-06
Exact statevector baseline for the kingston DLA-parity data — 2026-07-21catalogStatus: completed. Addresses audit findings B-9/B-10 (exact classical reference + mitigation-ablation context for the promoted DLA-parity data). Code: src/scpn_quantum_control/analysis/dla_parity_exact_baseline.py, runner scripts/ru
DLA State/Layout Randomisation PreregistrationcatalogDate: 2026-05-06
Entanglement Entropy / Tomography Check PreregistrationcatalogDate: 2026-05-06
GUESS / Symmetry-Decay Calibration PreregistrationcatalogDate: 2026-05-06
S1 Hybrid Classical-Quantum Feedback Loopprimary navDate activated: 2026-05-06
IBM Phase 2 Preregistered Run Manifest — 2026-05-05catalogThis manifest is the pre-submission contract for the next IBM hardware run. It is intentionally narrow: it permits a Phase 2 DLA parity expansion only, not a broad quantum-advantage claim and not any frontier or live-loop claim.
IBM Phase 2 B-C Scaling Manifest — 2026-05-05catalogThis manifest preregisters the next continuation run after the completed reduced A+G replication. It is limited to scaling blocks B and C only.
IBM Popcount-Control Manifest -- 2026-05-05catalogThis manifest preregisters the next QPU run. Its purpose is to test whether the observed n=4 DLA parity leakage asymmetry is primarily a parity-sector effect, primarily an excitation-count / amplitude-damping effect, or a combination of bot
IBM v2 raw-count recovery and full disclosure (2026-07-18)catalogThe "IBM v2" fair-experiment campaign (2026-03-29, ibm_fez) was originally committed aggregate-only — fidelity/survival means plus HMAC-blinded run labels, with no raw counts and no raw job identifiers. Because it could not be independe
IQM DLA Backend-Sensitivity Preregistration (Properly Powered)catalogDate: 2026-07-21
IQM Garnet Depth-10 Sign-Replication PreregistrationcatalogDate: 2026-07-22
IQM Garnet Parity-Asymmetry Depth-Profile PreregistrationcatalogDate: 2026-07-22
IQM DLA Submission-Journal RunbookcatalogDate: 2026-07-26
IQM Garnet Parity-Asymmetry Window-Variability Preregistration (FU-W)catalogDate: 2026-07-22
IQM Garnet Layout-Transfer Per-Size Powered Preregistration (FU-3)catalogDate: 2026-07-22
IQM FU-3 Per-Size Layout-Transfer ReadinesscatalogThis package implements the provider-free gates frozen in iqm_layout_transfer_per_size_prereg_2026-07-22.md. It does not authorize a provider call or hardware execution.
IQM Garnet FU-3 Per-Size Layout-Transfer ResultcatalogExecution date: 2026-07-26
IQM Square-Lattice Layout-Transfer PreregistrationcatalogDate: 2026-07-21
KYMA toy compositional-generalisation probe — 2026-07-18catalogStatus: completed, honest NEGATIVE result with a mechanistic diagnosis. Pre-registration (frozen, no metric shopping): .coordination/planning/CEO/KYMA_TOY_PROBE_PREREGISTRATION_7f6b_2026-07-18.md. Raw artifact: data/kyma_composi
KYMA v2.1 supplementary rigor — ablations, baselines, convergence, LOO — 2026-07-21catalogStatus: complete — all four analyses run. Strengthens and corrects the landed v2 PASS (db7c5c47). Code: src/scpn_quantum_control/benchmarks/kyma_v2/{ablations,baselines,rigor}.py, runner scripts/run_kyma_v2_rigor.py, tests tests
KYMA v2 composition probe — corrected design, PASS — 2026-07-21catalogStatus: run complete. Verdict: PASS (pre-committed contract, 5 seeds). Addresses the two defects diagnosed by the v1 NEGATIVE. Code: src/scpn_quantum_control/benchmarks/kyma_v2/, tests tests/test_kyma_v2_*.py, runner scripts/run
Larger-System Submission Extension Readinesscatalog- Generated: 2026-05-20T19:38:57+00:00 - Backend: generic_backend_127q (generic_estimator) - Hardware submitted: False - Ready QPU estimate, one backend: 1152.0s (19.20 min) - Ready QPU estimate, Fez+Marrakesh pair: 2304.0s (38.40 min) - IB
Larger-System Submission Extension Readinesscatalog- Generated: 2026-05-20T19:39:03+00:00 - Backend: ibm_fez (live_ibm_backend) - Hardware submitted: False - Ready QPU estimate, one backend: 1152.0s (19.20 min) - Ready QPU estimate, Fez+Marrakesh pair: 2304.0s (38.40 min) - IBM usage probe:
Larger-System Submission Extension Readinesscatalog- Generated: 2026-05-20T19:39:09+00:00 - Backend: ibm_marrakesh (live_ibm_backend) - Hardware submitted: False - Ready QPU estimate, one backend: 1152.0s (19.20 min) - Ready QPU estimate, Fez+Marrakesh pair: 2304.0s (38.40 min) - IBM usage
Larger-System Submission Extension RunbookcatalogGenerated: 2026-05-20.
Layer-Selective Qubit Assignment PreregistrationcatalogDate: 2026-05-06
Maximum-Width Kuramoto-XY Campaign Preregistration (WIDTH-1)catalogDate: 2026-07-16
Multi-Circuit QEC Demonstration PreregistrationcatalogDate: 2026-05-06
Next validation protocols -- no-submit preparationcatalogDate: 2026-05-05
On-QPU Dynamic-Circuit Feedback Demo Preregistration (RC-1)catalogDate: 2026-07-16
Phase 2 no-QPU cross-checkscatalogDate: 2026-05-05
Phase 3 Entanglement/Tomography IBM Execution PackagecatalogDate: 2026-05-20
Phase 3 Entanglement/Tomography Analysis ManifestcatalogDate: 2026-05-20
Phase 3 Entanglement/Tomography Analysis ManifestcatalogDate: 2026-05-20
Phase 3 Entanglement/Tomography Analysis ManifestcatalogDate: 2026-05-20
Phase 3 Entanglement/Tomography Paper PlancatalogDate: 2026-05-20
Phase 3 Entanglement/Tomography ReadinesscatalogDate: 2026-05-07
Phase 3 Kingston ZNE Statistical RepeatcatalogDate: 2026-05-20
Phase 3 Entanglement ZNE Analysis ManifestcatalogDate: 2026-05-20
Phase 3 Entanglement ZNE Analysis ManifestcatalogDate: 2026-05-20
Phase 3 Entanglement ZNE Analysis ManifestcatalogDate: 2026-05-20
Phase 3 Third-Backend ZNE Extension: ibm_kingstoncatalogDate: 2026-05-20
Phase 3 GUESS DLA ManifestcatalogDate: 2026-05-07
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_fez - Width: n=6 - Job IDs: d870qa8p0eas73dlj6kg - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T195233Z_ibm_fez_n6.json - Reference rows: data/phase3_entanglement_tomography/phase3_l
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_fez - Width: n=8 - Job IDs: d870qf9789is73909nu0 - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T195245Z_ibm_fez_n8.json - Reference rows: data/phase3_entanglement_tomography/phase3_l
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_marrakesh - Width: n=6 - Job IDs: d870qj9789is73909o2g - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T195300Z_ibm_marrakesh_n6.json - Reference rows: data/phase3_entanglement_tomogra
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_marrakesh - Width: n=8 - Job IDs: d870qo0p0eas73dlj75g - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T195312Z_ibm_marrakesh_n8.json - Reference rows: data/phase3_entanglement_tomogra
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_kingston - Width: n=6 - Job IDs: d871a3p789is7390a9p0 - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T201613Z_ibm_kingston_n6.json - Reference rows: data/phase3_entanglement_tomograph
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_marrakesh - Width: n=8 - Job IDs: d871aj1789is7390aa9g - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T201657Z_ibm_marrakesh_n8.json - Reference rows: data/phase3_entanglement_tomogra
Phase 3 Larger-System Analysis Manifestcatalog- Backend: ibm_kingston - Width: n=8 - Job IDs: d871bv1789is7390abng - Raw counts: data/phase3_entanglement_tomography/phase3_large_system_raw_counts_20260520T201817Z_ibm_kingston_n8.json - Reference rows: data/phase3_entanglement_tomograph
Phase 3 Layer-Selective Comparator MatrixcatalogDate: 2026-05-07
Phase 3 Layer-Selective Readiness ManifestcatalogDate: 2026-05-07
Phase 3 Multi-Circuit QEC ReadinesscatalogDate: 2026-05-07
Phase 3 Multi-Device DLA ManifestcatalogDate: 2026-05-06
Phase 3 Native Decomposition ReadinesscatalogDate: 2026-05-07
Phase 3 State/Layout DLA ManifestcatalogDate: 2026-05-07 Backend: ibm_marrakesh Jobs: ibm-run-aabcf620230b1438, ibm-run-eea172711aa52b78 Raw-count SHA256: 03068ddaa9794f1ac19614e700887a84dd013cd5af107f49b39c3cff9e5674ac
Phase 3 VQS Alternative ReadinesscatalogDate: 2026-05-07
QBER Re-Run With Committed Basis Metadata (KIMI-8)catalogDate: 2026-07-17
S6 Quantum-Kuramoto API ContractcatalogThis contract validates the proposed quantum_kuramoto export surface before any package skeleton is created.
S6 Quantum-Kuramoto Boundary ReviewcatalogThis review converts the import-graph audit into a conservative package-boundary decision. It does not create a quantum_kuramoto package skeleton.
Full-Basis Readout Calibration PreregistrationcatalogDate: 2026-05-06
Readout Full-Basis ManifestcatalogDate: 2026-05-06
Realtime Control E2E BenchmarkcatalogDate: 2026-05-22
Realtime Control Latency ConsolidationcatalogTimestamp: 2026-05-22T00:13:57.520930+00:00
Dedicated Realtime Latency CampaigncatalogTimestamp: 2026-05-22
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 7.144280 s (n=12) - Python mean: 6.789447 s (n=12) - Delta (Python - Rust): -0.354833 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 6.801363 s (n=6) - Python mean: 6.737755 s (n=6) - Delta (Python - Rust): -0.063607 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 6.517766 s (n=6) - Python mean: 6.409961 s (n=6) - Delta (Python - Rust): -0.107805 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 9.019362 s (n=6) - Python mean: 11.578785 s (n=6) - Delta (Python - Rust): 2.559423 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 6.928691 s (n=6) - Python mean: 7.126052 s (n=6) - Delta (Python - Rust): 0.197361 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 7.264898 s (n=6) - Python mean: 7.091577 s (n=6) - Delta (Python - Rust): -0.173320 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 8.305232 s (n=6) - Python mean: 7.180488 s (n=6) - Delta (Python - Rust): -1.124744 s
Realtime Control Latency Rust vs Pythoncatalog- Rust mean: 7.213458 s (n=6) - Python mean: 10.428570 s (n=6) - Delta (Python - Rust): 3.215112 s
S1 Feedback IBM Paper PlancatalogDate: 2026-05-20
S1 Feedback Readiness Indexprimary navDate: 2026-05-06
S1 IBM Feedback Pair ResultcatalogDate: 2026-05-20
S1 IBM Live ReadinesscatalogDate: 2026-05-20
S1 IBM Live ReadinesscatalogDate: 2026-05-20
S1 IBM Live ReadinesscatalogDate: 2026-05-20
S1 Live Submission Preflightprimary navDate created: 2026-05-06
S1b IBM Direct-XY Observable ResultcatalogDate: 2026-05-20
S1c IBM Shallow Direct-XY ResultcatalogDate: 2026-05-20
S1d IBM Policy-Direction Sweep ResultcatalogDate: 2026-05-20 Backend: ibm_kingston Experiment ID: s1d_policy_direction_sweep_2026-05-20 Parent experiment: s1_dynamic_feedback_preregistration_2026-05-06
S1e IBM Policy-Sweep Confirmatory RepeatcatalogDate: 2026-05-20 Backend: ibm_kingston Experiment ID: s1e_policy_sweep_confirmatory_repeat_2026-05-20 Parent experiment: s1_dynamic_feedback_preregistration_2026-05-06
S1f IBM Quadrature Mechanism CheckcatalogDate: 2026-05-20 Backend: ibm_kingston Experiment ID: s1f_quadrature_mechanism_check_2026-05-20 Parent experiment: s1_dynamic_feedback_preregistration_2026-05-06
S2 Full Scaling Campaign PlancatalogDate: 2026-05-07
S2 Full Scaling Campaign SlicecatalogDate: 2026-05-07
S2 Full Scaling Campaign SlicecatalogDate: 2026-05-07
S2 Full Scaling Campaign SlicecatalogDate: 2026-05-07
S2 Full Scaling Campaign SlicecatalogDate: 2026-05-07
S2 n=14 Resource GatecatalogDate: 2026-05-07
S2 Scaling Readiness Indexprimary navDate: 2026-05-06
S2 Slice Progress ReportcatalogDate: 2026-05-07
S3 Design Readiness Indexprimary navS3 starts as a no-QPU readiness gate for ML-augmented pulse and ansatz design. The current slice does not train a model and does not submit hardware jobs. It ranks deterministic candidate families so that later ML surrogate training has a v
S4 IBM pulse-level Kuramoto-XY calibration-review preregistrationcatalogJob ID: s4_ibm_pulse_calibration_review
S4 Multi-hardware ReadinesscatalogThis is a no-submit readiness artefact. It does not contact cloud providers, reserve QPU time, or authorise hardware execution.
S4 neutral-atom Kuramoto-XY provider-object preregistrationcatalogJob ID: s4_neutral_atom_provider_object_review
SCPN/FIM Claim Boundaryprimary navDate: 2026-05-05
SCPN/FIM Self-Referential Hamiltonian Paper Planprimary navDate: 2026-05-05
SCPN/FIM IBM pilot analysiscatalogDate: 2026-05-05
SCPN/FIM repeated IBM follow-up analysiscatalogDate: 2026-05-05
SCPN/FIM repeated IBM follow-up protocolcatalogDate: 2026-05-05
SCPN/FIM Hamiltonian Paper Submission ChecklistcatalogDate: 2026-05-06
SCPN/FIM Validation Protocolprimary navDate: 2026-05-05
Two-edge-colour width-2 schedule for the XY-Trotter chain — 2026-07-21catalogStatus: design + classical evidence complete. Addresses audit AUD-7 (B-5): hand-scheduled, genuinely width-2 chain dynamics instead of a deep serial synthesis. The hardware submission remains an owner-gated QPU run (AUD-5/7). Code:
Validation Readiness Gate — 2026-05-05catalogThis gate records what is ready for new validation runs and what remains blocked. It exists to prevent aggregate-only, placeholder, queued-job, or fallback-count artefacts from being promoted as hardware validation.
Variational Quantum Simulation Alternative PreregistrationcatalogDate: 2026-05-06

Learning and guides

PageSurfacePurpose
Differentiable Tutorialsprimary navThis page is the practical differentiable-programming path. It connects the physics object, support diagnostics, framework boundaries, compiler report, and training evidence without requiring provider credentials or hardware access.
Example Galleryprimary navThirty task-shaped entry points, ordered from the fastest no-credential run to hardware-evidence, integration, and release-readiness surfaces. Each entry states what it does, when to reach for it, and the fastest safe command. Every command
Hardware Execution Guideprimary navThe hardware package provides the full stack from circuit compilation to QPU execution, noise modelling, classical reference computation, and multi-backend support. 17 modules (April 2026: added qubit_mapper.py for DynQ topology-agnostic pl
PennyLane + Qiskit migration guides product (BL-41)catalogVersioned adoption-path product mapping PL/Qiskit concepts to SCPN APIs with materialised local round-trips and honest Runtime boundaries.
QNN, QGNN, and QSNN convergence examples (BL-42)primary navscpn_quantum_control.ml_examples provides one deterministic, simulator-only training example for each of the repository's QNN, QGNN, and QSNN model families. The suite composes the existing trainers; it does not introduce a second optimisat
Complete notebook catalogprimary navThis page indexes every committed notebook without rewriting notebook cells. Use the curated interactive notebook guide for the recommended learning sequence and this catalog when you need a specific experiment.
Differentiable notebook programme product (BL-40)catalogVersioned core-six onboarding curriculum under notebooks/differentiable/ with mandatory hardware_execution: false honesty.
Interactive Notebooksprimary nav*109 tracked Jupyter notebooks covering the full journey from basic Kuramoto dynamics to frontier research. Notebooks 01-13 cover core quantum simulation. Notebooks 14-47 document the FIM (Fisher Information Metric) strange loop investigati
Tutorials and Learning Pathprimary navThis page guides you from your first simulation to publishing-quality research results. Each section builds on the previous, progressing from conceptual understanding through hands-on computation to original research.

Literature

PageSurfacePurpose
Literature Catalogueprimary navThis directory stores project-relevant literature surveys and citation-planning material. Keep literature reviews here instead of the repository root.
Literature Survey: Quantum Computing Advances Relevant to scpn-quantum-control (2024--2026)primary navDate: 2026-03-20 Scope: NISQ quantum simulation of coupled Kuramoto oscillators mapped to XY Hamiltonians on IBM hardware Purpose: Research planning for v1.0 hardware campaign and arXiv preprint

Operations and governance

PageSurfacePurpose
Hardware Result-Pack Release Checklistprimary navThis checklist is mandatory for any tag, paper-facing update, website update, or release note that cites promoted IBM hardware evidence. It is intentionally offline: the release process must preserve committed artefacts and replay count-to-
Licensing FAQprimary navThis page states the current licensing boundary for scpn-quantum-control. It is documentation for adopters and release reviewers; it does not change any licence grant.
Release Readiness Gateprimary navDate added: 2026-05-18

Publication

PageSurfacePurpose
Publication Operationsprimary navThis directory contains submission checklists, source-packaging notes, registry plans, and venue-readiness material.
arXiv Source Packaging Readinessprimary navDate prepared: 2026-05-06
Combined Paper Submission ChecklistcatalogDate: 2026-05-06
Conda-forge SubmissioncatalogDate submitted: 2026-05-06
DLA Parity Preprint Submission ChecklistcatalogDate: 2026-05-06
JOSS-Style Software Paper Submission ChecklistcatalogDate: 2026-05-06
Metriq Local SmokecatalogDate: 2026-05-06
Metriq SCPN Benchmark Schema ProposalcatalogDate: 2026-05-07
Metriq Submission DecisioncatalogDate: 2026-05-07
Metriq Submission ReadinesscatalogDate checked: 2026-05-06
Phase 2 Publication Package -- 2026-05-05catalogThis manifest defines the release-ready Phase 2 publication package for the DLA parity hardware continuation. It is an index of committed artefacts only; it does not introduce new analysis results.
pyOpenSci Submission Readinessprimary navDate prepared: 2026-05-06
Qiskit Ecosystem SubmissioncatalogDate submitted: 2026-05-06
Registry Listing PlancatalogDate: 2026-05-06
Rust/VQE software-methods paper plancatalogDate: 2026-05-05
Rust/VQE Methods Paper Submission ChecklistcatalogDate: 2026-05-06
SciPy 2026 CFP Readinessprimary navDate prepared: 2026-05-06
Software Heritage Archive RecordcatalogDate checked: 2026-05-06

Science and engineering

PageSurfacePurpose
Architecture Decision Recordsprimary navThis page records the major design decisions behind scpn-quantum-control: for each one, the context that forced a choice, the decision taken, why it was taken, the alternatives weighed, and the trade-offs accepted. It complements Architectu
Experiment Roadmap: March-June 2026primary nav> Canonical status note (2026-05-06): active experiment selection > is consolidated in ROADMAP.md under "Canonical work queue". This > document remains the detailed historical experiment ledger and should > not independently authorise Q
Export Control AssessmentcatalogNot legal advice. This document is the project's own good-faith analysis of which export-control regimes apply to scpn-quantum-control, performed on 2026-04-17. A consumer planning to redistribute, embed, or combine this code with propr
Paper Claims: Quantum Simulation of Kuramoto Phase Dynamics on NISQ Hardwareprimary navThis file is a legacy planning and triage note, not a submission-ready claim source. Use the current manuscripts, hardware ledger, raw-count artefacts, and claim-boundary documents for coauthor review and publication wording. Items below pr
Quantum Cryptography Research Branchprimary navThis branch records experimental work on topology-authenticated cryptography built on the SCPN coupling matrix model. It is a specialist research lane and is separate from the promoted default API and release gates.
REPRODUCIBILITY RECORD — SCPN Quantum Control Frontier Campaign 2026catalogDocument version: 2.0 Date: 2026-04-26 Author: Miroslav Šotek / SCPN Quantum Control project SPDX: AGPL-3.0-or-later
Threat ModelcatalogThis document names the things we are trying to protect (assets), the kinds of actors we plan against (adversaries), where they can try to interact (attack surface), and the mitigations currently in the repository. It is a companion to SECU
Actions History Dashboardprimary navThis page documents the read-only workflow-history audit layer used to keep GitHub Actions history interpretable without deleting evidence for unresolved defects.
Adaptive Branching ReadinesscatalogThis is the S8 no-submit readiness surface for mid-circuit adaptive branching. It records branch policies and backend prerequisites, with no hardware submission and no adaptive-advantage claim.
Adaptive FIM next-experiment proposalsprimary navBL-80 turns leakage and exact-state-retention counts into conservative proposals for a future static lambda_fim batch. It is an offline experimental-design surface, not a realtime controller and not evidence that FIM feedback protects hardw
Advantage / no-advantage language protocol (BL-65)catalogThis page is the operator-facing guide for when “advantage” language is allowed. Default posture is no-advantage / research observation. Decisive protocol modules remain claim-gated evidence paths — never invent-green marketing.
Analog Kuramoto Backendsprimary navThe analog Kuramoto backend interface compiles a validated K_nm, omega problem into native programme schemas for three hardware families:
Analog Oscillator Mapping FeasibilitycatalogBL-35 provides a fail-closed answer to a deliberately narrow question:
Analog-Native Kuramoto ReadinesscatalogThis is the S10 no-submit readiness surface for analog-native Kuramoto backends. It records primitive accounting and provider export status without hardware submission or analog-advantage promotion.
Architectureprimary navThis page documents the software architecture in a way that supports technical due diligence and long-lived integration. The design objective is to keep problem-to-experiment flow deterministic while allowing each subsystem to evolve with c
Architecture Map — capabilities, IO, backends, wiringprimary navThis page is the capability/input-output/backend map of scpn-quantum-control, written for sibling repositories (sc-neurocore, scpn-control, scpn-fusion-core, scpn-phase-orchestrator) and for STUDIO design. It complements two existin
Artifact Storage Auditprimary navThis audit covers committed files under results/, data/, and figures/ as of 2026-04-29. It exists to decide whether the current artifact footprint belongs in Git, Git LFS, or a Zenodo/download-only route.
Attested result packs (BL-48)catalogStrip-resistant content digests bound to claim axes — never a self-asserted “validated” badge. Local unsigned envelopes are first-class; absent keys yield UNGRADED, never silent-validated.
Automatic Backend Selectioncatalogscpn_quantum_control.phase.backend_selector
Backends & Plugin RegistrycatalogTwo modules for runtime backend management:
Benchmark: Dynamic Coupling (Quantum Hebbian Learning)primary navModule: scpn_quantum_control.qsnn.dynamic_coupling Class: DynamicCouplingEngine
Benchmark: Hardware Topological Optimizerprimary navModules:
Benchmark: Structured Ansatzprimary navModule: scpn_quantum_control.phase.structured_ansatz Function: build_structured_ansatz
Benchmark HarnesscatalogThis page provides the reproducible entry point for public benchmark reconstruction. It keeps low-cost datasets, loaders, and tolerance checks aligned so published benchmark rows can be regenerated without submitting jobs.
Changelogprimary navThis file is the release history anchor for API, evidence, and policy changes. Use it to verify when a claim boundary, verification path, or gating rule changed before that version was promoted.
Chimera and Multiscale Synchronisation Controlprimary navBL-60 turns finite synthetic chimera states, nested order parameters, and hierarchical coherence targets into a documented local control workflow. The implementation composes the repository's existing Kuramoto–Sakaguchi force, Shanahan diag
Choosing a gradient pathprimary navThis is the front door for the differentiable-programming surface: a fast route from what you are differentiating to which entry point to call. It is a guide, not a promotion — see Claim status before you quote any number.
Classical Baselinesprimary navThis page records the supported classical baseline surfaces for Kuramoto-XY workflows. Each baseline returns a ClassicalBaselineRun envelope with the backend name, availability flag, elapsed wall time, time grid, order-parameter trajectory,
Classical irreproducibility of the DLA-parity asymmetryprimary navThe DLA-parity hardware campaign on ibm_kingston measured a non-zero asymmetry between the "even" and "odd" initial-state sectors: peak +17.48 % at depth 6, mean +9.25 % across the eight depths surveyed, with Fisher combined $\chi^{2} = 123
Closed-Loop Control Analysisprimary navscpn_quantum_control.control.closed_loop_analysis turns the response of the measurement-feedback synchronisation controller (control.realtime_feedback.RealtimeSyncFeedbackController) into a control-theoretic verdict and gates any non-simula
Continuous competitive baseline watch (BL-61 / W1)catalogFail-closed ops watch over differentiable competitive baselines. Composes the committed refresh inventory (differentiable_competitive_baselines) into a queryable catalogue with pin / version / refresh honesty and structured feed probes
Control Scope Boundaryprimary navThe scpn_quantum_control.control package is scoped to synchronisation-control workflows that already exist in this repository:
Core Package Boundaryprimary navThis page records the intended boundary for a possible future lightweight Kuramoto-XY core package. It is a planning boundary only. All code currently published in this repository remains under AGPL-3.0-or-later, with the commercial licence
Count-Integrity Incident, April 2026primary navPublished 2026-07-16. This note makes the April 2026 internal incident record public in one place; the canonical promotion status of every artefact remains the hardware status ledger.
Datasets and Data Registrycatalogscpn-quantum-control ships two classes of data artefact:
Differentiable External-Validation Technical Reportprimary navThis report defines the public comparison and reproducibility package for the differentiable-programming lane. It is a technical report, not a promotion claim. Every row remains bounded by the committed claim ledger until external compariso
Differentiable Programmingprimary navscpn-quantum-control treats differentiability as a product surface, not a hidden implementation detail. The goal is to make coupled-oscillator quantum control trainable, testable, and explainable across quantum gradients, classical program
Differentiable Reviewer Evidenceprimary navThis page maps the bounded DP-015 reviewer criticisms and the DP-030 evidence package to real repository commands, artefacts, and explicit public gaps. It is an index into executable evidence, not a replacement for the tests or artefacts.
Differentiable Roadmapprimary navThis roadmap defines the staged work needed to turn differentiable programming into a complete public product surface. It complements the private execution tracker without exposing private execution notes.
Differentiable Support Matrixprimary navThis page is generated from executable registry and planner surfaces. It is checked against the committed capability manifest so source modules, public registry exports, focused tests, and this documentation page cannot drift independently.
DLA-parity validation pathwayprimary navThe scpn_quantum_control.dla_parity subpackage is the open-data validation surface for the DLA-parity campaign on ibm_kingston. It loads the 342 published circuits, recomputes every published scalar from the raw counts, cross-checks each on
DLA-Protected Scar Memoryprimary navThe scar-memory prototype uses the DLA parity theorem and the fixed-parity repetition-code memory sector from qec.dla_protected_subspace. It prepares a logical cat state across two synchronised repetition-code words inside one DLA parity se
DLA-Protected Logical Synchronisationprimary navThe DLA-protected logical synchronisation module builds a finite repetition-code memory sector inside the global parity decomposition of the heterogeneous Kuramoto-XY Hamiltonian.
DLA and Topology-Constrained Differentiable Controlprimary navBL-54 exposes a bounded local derivative surface for two different constrained objects:
Complete documentation catalogprimary navThe primary navigation stays intentionally compact. This catalog keeps every public guide, evidence note, contract, campaign protocol, and reference page discoverable without crowding the main learning path.
DynQ: Dynamic Topology-Agnostic Qubit MappingcatalogQuality-weighted community detection for intelligent qubit placement on NISQ hardware. Replaces Qiskit's default SABRE-based layout heuristic with a physics-aware global partitioning strategy.
E2E contract boundariesprimary navThis page records the current end-to-end and contract-test boundary coverage used for release-safety review. The audit is intentionally conservative: a boundary is marked covered only when a pytest module name maps to that boundary, and st
End-to-End Testingcatalogtests/test_e2e_new_modules.py
ENAQT Optimal-Noise Scanprimary navBL-87 provides a bounded local simulator for environment-assisted quantum transport (ENAQT). It scans a finite set of local dephasing rates and maximises the population irreversibly transferred into a target sink by a fixed time. The commit
Entangled initial-state coherence studyprimary navBL-79 is a bounded four-qubit simulation study of how four pure initial-state families evolve under one frozen Kuramoto-XY Hamiltonian. It does not show that entanglement lowers a synchronisation threshold. The model is closed and unita
Quantum Random-Number Generationprimary navThe scpn_quantum_control.entropy package provides a streaming quantum random-number generator with the NIST SP 800-22 Revision 1a statistical test suite and the FIPS 140-2 Annex C power-up tests. Entropy originates from Qiskit Aer measureme
Key Equationsprimary navThis page collects every equation implemented in scpn-quantum-control, from the foundational Kuramoto-to-qubit mapping through the 33 research gems that probe the synchronization transition. Each equation is accompanied by the module that i
Error Mitigationcatalogscpn-quantum-control ships three complementary error-mitigation backends. Choose based on whether you have a known symmetry observable, how much shot budget you can spend, and whether you want a generic black-box wrapper.
Execution-Surface Policycatalogscpn-bench executes fixed repository scripts with local user privileges. Each harness therefore carries an explicit ExecutionSurfacePolicy:
Falsification ProtocolcatalogA scientific claim is only meaningful if there is an experiment whose outcome would refute it. This page collects the falsification criteria for every non-trivial claim scpn-quantum-control currently makes, so a reader can locate the break
FRC Pulsed-Shot QAOA Schedulingprimary navscpn_quantum_control.control.qaoa_pulsed_cost and scpn_quantum_control.control.frc_pulsed_qaoa schedule capacitor-bank firing for a field-reversed-configuration (FRC) pulsed-compression shot by minimising a control-grade physics cost with Q
Free Energy Principle (FEP) on Quantum SubstratecatalogThe Free Energy Principle (Friston 2010) defines variational free energy as:
Governed multi-ecosystem route matrix (BL-52)primary navThis page is the operator-facing guide for the fail-closed multi-ecosystem route matrix productised under BL-52. It answers: which differentiable route IDs exist, what their closure status is, and why alternatives were rejected — with
GPU Batch VQEcatalogscpn_quantum_control.phase.gpu_batch_vqe
Reproducible Hardware Result Packsprimary navHardware result packs are the repository's offline trust surface for promoted IBM hardware evidence. A pack binds raw counts, reproduced summaries, IBM job identifiers, byte sizes, SHA-256 digests, reproduction commands, and explicit claim
Hardware-safe gradient execution product (BL-47 / Axis 5)catalogFail-closed no-submit default execution policy surface for hardware-adjacent gradient planning. Dry-run plans estimate shot budgets and cost-model status without provider submission; enforce refuses would-submit and over-budget paths.
Hardware Status Ledgerprimary navResult-pack integrity layer: promoted IBM raw-count datasets are now indexed in hardware_result_packs.md, with the canonical manifest at data/hardware_result_packs/manifest.json and an offline verifier at scripts/verify_hardware_result_pack
Hybrid Digital-Analog Executionprimary navThe hybrid compiler splits a validated Kuramoto-XY workload into two coherent execution blocks:
Isolated Benchmark Runnerprimary navDifferentiable-benchmark evidence is promoted to isolated_affinity only when it runs on a self-hosted GitHub Actions runner with a reserved CPU, a recorded governor and frequency, and low host load. Without such a runner the differentiable-
JAX NQS baseline productcatalogscpn_quantum_control.jax_nqs_baseline_product is the BL-103 exact-reference evidence layer for the ambient JAX restricted-Boltzmann-machine runner. It is deliberately a small-system research baseline.
Josephson K_nm Magnitude Studyprimary navThis QWC-5.2 artifact records the Josephson topology-correlation candidate and the measured-magnitude gates required before any physical K_nm coupling claim.
Kuramoto Competitive Benchmarkprimary navThis page documents the external competitive harness that measures our Kuramoto toolkit against real third-party solvers on one deterministic Kuramoto forward problem. It is the cross-package counterpart to the in-repository Kuramoto Tier B
Kuramoto competitive evidence — oscillatools vs the Julia SciML toolsprimary navThis note records a measured head-to-head between the oscillatools coupled-phase-oscillator toolkit and the established Julia libraries on two axes — integration throughput and differentiability — so the toolkit's standing is stated
Kuramoto Core Facadeprimary navThe scpn_quantum_control.kuramoto_core facade is the stable entry point for users who only need the Kuramoto-XY compiler layer:
Kuramoto Handbookprimary navThis handbook is the reference page for the in-repository Kuramoto toolkit. It is generated from the live scpn_quantum_control.kuramoto facade and the committed multi-tier benchmark artefact, so the public API inventory, model families, and
Kuramoto JAX delayed-tierprimary navThe time-delayed Kuramoto model θ˙j(t)=ωj+kKjksin(θk(tτ)θj(t))\dot\theta_j(t) = \omega_j + \sum_k K_{jk}\sin(\theta_k(t-\tau) - \theta_j(t)) is integrated by accel.kuramoto_delayed.integrate_delayed_kuramoto, a delay-aware method-of-steps RK4, and its gradient by the
Kuramoto JAX tierprimary navThe networked-Kuramoto production integrators dispatch Rust → Julia → NumPy. This page documents the opt-in JAX tier: fixed-step Euler, RK4, adaptive Dormand-Prince, networked inertial RK4, networked symplectic inertial, and seeded nois
Kuramoto JAX differentiable-model MPC tierprimary navModel-predictive control (MPC) plans a finite-horizon control from the current measured state, applies the first control, then re-plans from the new measurement. This page documents an MPC for the Kuramoto network whose **predictive model i
Sparse Kuramoto CPU Pathprimary navoscillatools exposes an explicit sparse CPU route for large classical Kuramoto networks:
Kuramoto Standalone Package Decisionprimary navDecision status: APPROVED 2026-07-04 (CEO/IP). This record supersedes the prior deferral (2026-06-26) and satisfies the Promotion Gate by recording all eight required decisions below. Implementation proceeds in phases; the standalone packag
Higher-Order, Monitored, and PT-Symmetric Kuramoto Variantsprimary navscpn_quantum_control.phase.kuramoto_variants formalises three Kuramoto extensions that previously only existed as campaign-level experiments:
Kuramoto Visualisation Layerprimary navscpn_quantum_control.accel.kuramoto_visualisation (re-exported from the scpn_quantum_control.kuramoto facade under the visualisation capability group) renders the standard diagnostics for coupled-phase-oscillator trajectories directly from
Language PolicycatalogThis document codifies which programming language this repository uses for which part of the stack, and under what criteria a module picks a particular language. It exists so that every new module has a defensible, written rationale for its
KT-4 — Continuous-Relaxation Layout Search: Research Design (RESEARCH LABEL)catalog**Status: RESEARCH — ANSWERED (2026-07-16). The preregistered experiment ran (§6): the null hypothesis stands — the relaxation shows no gain over the KT-3 discrete baseline at matched true-cost budget. The research label stays; the discrete
Lindblad Master Equation Solvercatalogscpn_quantum_control.phase.lindblad
Logical DLA Parity RoadmapcatalogThis note is a post-NISQ planning artefact. It estimates resources for logical-level DLA parity work and keeps the survival claim blocked until the theory and simulation prerequisites are closed.
Metamorphic AD verification catalogue (BL-46)catalogThis page is the operator-facing guide for gradient correctness beyond one-off examples: a versioned catalogue of metamorphic laws, fail-closed refuse paths (including invent-green hardware formal proofs), and pure residual band checks.
Methodology and Evidence Governanceprimary navThis is the single entry point for the project's evidence-governance material. You do not need to read it to run the examples — start with Onboarding and the Example Gallery. Read this page when you need to understand *what a result in this
Methods Benchmark Dashboardprimary navThis page is the public reproducibility dashboard for the benchmark artefacts supporting the Rust/VQE methods papers and the SCPN/FIM Hamiltonian paper. The rule is artefact-first: tables and manuscript claims should be regenerated from com
Post-Quantum Trigger Signer (ML-DSA-65)primary navscpn_quantum_control.crypto.ml_dsa is a from-specification implementation of the FIPS 204 ML-DSA-65 module-lattice digital signature scheme, and scpn_quantum_control.crypto.pqc_trigger builds a capacitor-bank trigger authorisation signe
Multi-Platform Circuit Exportcatalogscpn_quantum_control.hardware.circuit_export
Multimodal Forecasting Under Partial Observationprimary navThis page defines the bounded BL-37 multimodal forecasting product. It provides immutable multimodal custody, a deterministic missingness-aware classical baseline, partial-observation scoring, empirical split residual intervals, and explici
Multi-Scale Quantum Error Correction (MS-QEC)catalogMS-QEC implements hierarchical quantum error correction through concatenated surface codes. Each SCPN domain operates an independent surface code whose logical qubits serve as the physical qubits of the next domain's code.
Mutation TestingcatalogLine-coverage tells you which statements ran. It does not tell you which logic changes your tests would catch. Mutation testing flips operators and constants in the source one at a time, reruns the suite, and measures how many mutants "
Native Speedup Benchmarkprimary navThis page documents the reproducible Rust-vs-Qiskit benchmark for dense XY-Hamiltonian construction — the operation the retired "5401× faster than Qiskit" headline referred to. It follows the unified GOTM benchmark standard (agentic-shared/
Kuramoto neural-operator advantage studyprimary navThis page reports whether the DeepONet neural-operator surrogate for Kuramoto network dynamics (the forecasting.kuramoto_neural_operator module) does more than reduce its training loss — whether it forecasts unseen initial conditions accura
Neural Quantum StatescatalogTwo modules for variational ground state search using neural network wavefunctions:
New Module IndexcatalogThis page preserves the historical March 2026 module notes and now routes the newer v0.10 public surfaces to their maintained guides. Use the guide pages below for current examples, evidence boundaries, and API contracts before reading the
NV-Centre 20 T Magnetometryprimary navscpn_quantum_control.sensing.nv_magnetometry_20T is a simulation-only nitrogen-vacancy (NV) magnetometry response model valid into the 20 T regime. The ground-state spin-1 Hamiltonian is diagonalised exactly, so the optically-detected m
Open-System Hardware CircuitscatalogHardware-executable methods for simulating open quantum systems on NISQ devices. Two complementary approaches:
Orchestrator Integrationprimary navThis page defines the interoperability boundary between scpn-quantum-control and scpn-phase-orchestrator. It is the source of truth for how phase and control semantics move from orchestrator specs into executable quantum workflows without d
p_h1 Open-Claim Guardprimary navThe QWC-5.3 guard keeps the p_h1 = 0.72 threshold publicly framed as an open empirical/theoretical parameter.
Dynamical Lie Algebra Parity Theorem for XY-Coupled Oscillator Networksprimary navMiroslav Šotek — ANULUM / Fortis Studio ORCID: 0009-0009-3560-0851
Synchronisation Witness Operators for Quantum Oscillator Networksprimary navMiroslav Šotek — ANULUM / Fortis Studio ORCID: 0009-0009-3560-0851
Physics-First Kuramoto-XY Tutorialprimary navThis tutorial starts from the physics object: a network of coupled oscillators. It does not require IBM credentials, SCPN layer constants, or a hardware account. The goal is to move from K_nm and omega to a Hamiltonian, a Trotter circuit, a
Pipeline Performance Benchmarksprimary navEvery module in scpn-quantum-control is verified as wired into the pipeline (not decorative) by tests/test_pipeline_wiring_performance.py (155 tests) and per-module pipeline tests embedded in each test file. This page documents the meas
Pipeline Runtime Contractprimary navEvery component that claims to participate in the SCPN stack must expose a runnable pipeline path. Importable modules are not enough: each pipeline needs a command, a typed contract, validation, and an auditable output artifact.
Bit-exact polyglot parity certificates (BL-49 / P1)catalogVersioned externally checkable certificate product for the Rust Program AD replay moat: family catalogue (scalar → spectral bounds), certificate schema, digest build/verify helpers. Digests prove sample bundle identity — they do not
Quantum Simulation of Coupled-Oscillator Synchronisation on a 156-Qubit Superconducting Processorprimary navMiroslav Šotek ANULUM / Fortis Studio, Marbach SG, Switzerland protoscience@anulum.li | ORCID: 0009-0009-3560-0851
Pre-registration Protocol for Hardware CampaignscatalogHardware experiments on a shared QPU are expensive and under-specified by nature — the submitter decides post hoc what counts as a "successful" run. Without a protocol frozen before the first circuit is submitted, hypothesis-after-the-resul
Rust Program AD fuzz assurance (BL-96 / P1)catalogVersioned fuzz trust-moat product over ambient scpn_quantum_engine/fuzz cargo-fuzz bins: target catalogue, time-boxed CI-optional policy, and dry-run probe helpers. Does not execute cargo-fuzz or invent-green continuous multi-hour c
Ψ-field Lattice Gauge SimulatorcatalogThe Ψ-field is modelled as a compact U(1) gauge theory on the SCPN graph. Link variables Uij=eiAijU_{ij} = e^{iA_{ij}} live on edges, where Aij[π,π)A_{ij} \in [-\pi, \pi) is the gauge potential.
QEC Decoder Boundaryprimary navThis page records the shipped quantum-error-correction decoder surfaces and the decoder families that are intentionally not claimed.
QPU Compute Unit Architectureprimary navThe QPU is not only a validation target for isolated tests. In the SCPN stack it should become a scheduled compute unit: a probabilistic accelerator that accepts typed oscillator artifacts, runs selected quantum kernels, and returns auditab
QPU Data Artifact Contractprimary navThe QPU data artifact is the publication gate between source-facing repositories and scpn-quantum-control. It prevents campaign scripts from inventing coupling matrices locally and makes every QPU submission traceable back to a source, repl
QPU Provider Readiness Matrixprimary navThis document prepares the SCPN stack for allocation requests across known accessible QPU families. The goal is not to hard-code one vendor; the goal is to make every provider an instance of the same compute-unit contract:
QPU System Capability Dossierprimary navThis dossier states what data the SCPN quantum-control stack needs, what computations it can route to quantum hardware, how the results are used, and what the system can realistically deliver. It is written for engineering execution and fut
Quantum-classical co-design loop (BL-33)primary navscpn_quantum_control.codesign composes the existing phase-objective, gradient-planning, open-system, control-adapter, and observer products into a deterministic local loop. It is a software-in-the-loop research surface. It does not submit p
Quantum/Classical Co-Simulation of K_nm Networksprimary navscpn_quantum_control.cosimulation simulates a large Kuramoto-XY coupling network by evolving a small, strongly-coupled core as an exact quantum statevector while the weakly-coupled remainder runs as a classical Kuramoto bath. A network of N
Quantum Gradientsprimary navQuantum gradients are the first differentiable-programming surface that most quantum-ML users look for. The current public route starts with parameter-shift gradients and expands toward backend-aware gradient planning, stochastic finite-sho
Bounded Quantum Graph Neural Networkprimary navscpn_quantum_control.phase.qgnn is a local quantum graph neural network that maps a K_nm coupling graph to a registered Phase-QNode circuit output. A classical message-passing stack turns the graph (nodes carry natural frequencies and weigh
Quantum Reservoir Computing and Classical Surrogatesprimary navThis page defines the bounded BL-45 quantum reservoir computing (QRC), matched classical baseline, and differentiable classical-surrogate surfaces.
Quantum Sensing ReadinesscatalogThis is the S11 no-submit readiness surface for DLA-driven quantum sensing via the synchronisation order parameter. It records QFI and classical Fisher proxy rows without hardware submission or sensing advantage promotion.
Quantum Thermodynamics ReadinesscatalogThis is the S9 no-submit readiness surface for thermodynamic signatures of synchronisation transitions. It records calibrated observables and protocol prerequisites with no hardware submission and no thermodynamic peak claim.
Real-Data Synchronisation Forecastingprimary navThe forecasting benchmark evaluates whether an early observed synchronisation window can calibrate a physical Kuramoto forecast without seeing the held-out samples. It is intentionally small and replayable: the default dataset is the commit
Real-Time Feedbackprimary navThe real-time feedback controller closes the Kuramoto-XY loop in software while keeping the circuit shape compatible with dynamic-circuit execution. It is not a hardware latency claim: the shipped path is a deterministic statevector/live-sh
Real-Time Runtimeprimary navscpn_quantum_control.control.realtime_runtime provides software timing contracts for bounded control loops. It has two surfaces:
Research Gems: Novel Quantum Probes of Synchronizationprimary navA legacy curated set of 33 research-gem modules at the intersection of quantum information, condensed matter physics, and the SCPN coupling architecture.
Deep-analysis research lanesprimary navSCPN Quantum Control contains a substantial collection of analysis and gauge modules. They do not all have the same maturity, evidence, or relationship to differentiable control. The BL-84 registry makes those differences explicit without t
Compile & dense resource budget gate (BL-94 / W5)catalogFail-closed resource budgets for sparse Pauli/compile construction and dense Hilbert-space allocations. Product catalogue + probe over the low-level guards in compile_budget and dense_budget (same estimate formulas; no silent diverging
Resultsprimary navKuramoto-XY simulator, compiler, and hardware-evidence ledger for heterogeneous-frequency coupled oscillators.
RL research governanceprimary navBL-102 keeps reinforcement-learning-adjacent code in an explicit, reproducible research lane. It does not turn witness search or pulse optimisation into a production controller.
Rust Acceleration Engineprimary navscpn_quantum_engine is an optional PyO3 extension module that accelerates hot-path computations. All Python modules transparently fall back to pure Python/NumPy when the Rust engine is not installed.
Scorecard acceptance engine (BL-56 / W1)catalogFail-closed promotion surface for the eleven differentiable baseline-scorecard categories. Honest behind_baseline inventory is the default until claim-ledger and external-comparison evidence packages exist.
Sparse Hamiltonian Constructioncatalogscpn_quantum_control.bridge.sparse_hamiltonian
Stable Core Backend Capability Matrixprimary navThis generated page records stable backend capability profiles.
Stable Core Contract FixturescatalogThese no-QPU, no-network fixtures lock stable core contract payloads.
Stable Core Preflight Fixturesprimary navThese no-QPU, no-network fixtures lock stable core preflight branches.
HAL Status Normalisation PolicycatalogThis document defines the frozen canonical status contract for hardware HAL adapters. Any alias change is a contract change and must follow the evidence gate below.
Strategic Roadmap — Post-v1.0 Differentiationprimary nav> Canonical status note (2026-05-06): active task selection is > consolidated in ROADMAP.md. This document remains the detailed > deferred/CEO-gated strategic catalogue; items here are not activated > unless copied into the canonical wo
Studio Executive Actionsprimary navThe QUANTUM studio is an executive tool, not only a federation publisher. The Studio Federation surface is the informative layer the SCPN-STUDIO hub ingests; the executive spine described here is the layer that actually runs a verb,
Studio Federationprimary navThe QUANTUM Studio federation surface has two layers:
Symmetry-Aware Exact DiagonalisationcatalogThree modules exploit symmetries of the XY Hamiltonian to reduce the Hilbert space dimension for exact diagonalisation (ED):
GUESS: Guiding Extrapolations from Symmetry DecayscatalogPhysics-informed zero-noise extrapolation using Hamiltonian symmetry observables to guide the mitigation of target observables on NISQ hardware.
Symmetry- and Sector-Aware Mitigation Compilerprimary navThis page defines the bounded first contract for the mitigation compiler lane. The current implementation is a planner, not a circuit transformer. It decides whether existing primitives are eligible for a Kuramoto/XY experiment descriptor a
Symmetry-Sector Mitigation Planner Fixturesprimary navThese no-QPU fixtures lock the planner contract before execution-path integration.
Kuramoto Chain n=8 Synchronisation Benchmarkprimary navThis no-QPU artefact records schema-compatible reference rows for the canonical eight-node Kuramoto chain with decaying coupling.
Kuramoto Ring n=4 Synchronisation Benchmarkprimary navThis no-QPU artefact records schema-compatible reference rows for the canonical four-node Kuramoto-XY ring benchmark.
Standardised Synchronisation Benchmark Suiteprimary navThis suite defines canonical coupled-oscillator benchmark instances and a stable result schema. It is a contract for future backends, not a new hardware claim.
Tensor Network MethodscatalogTwo modules for tensor-network-based simulation of the Kuramoto-XY system:
Test Infrastructureprimary navscpn-quantum-control has a CI-gated test suite with a 90% aggregate line coverage requirement. CI now collects branch arcs in the same run and requires real branch opportunity data, but the branch percentage remains observational until
Theoretical Foundationsprimary navThe Scale-Coupled Phase Network (SCPN) and its quantum simulation.
Theory-Hook Promotion Matrixprimary navBL-98 distinguishes a useful research routine from an admitted product or control capability. Importability is not promotion. Every reviewed hook has a machine-readable tier, one permitted role, explicit forbidden claims, and a small local
Multi-language Kuramoto tier benchmark — CI vs localprimary navPer-call P50 latency (microseconds) of every registered Kuramoto compute primitive across the Rust, Julia, and Python tiers, measured by scripts/bench_kuramoto_tiers.py. The canonical column is the fixed CI runner (Rust engine built fro
TN/MPS Baseline Designprimary navThis is the QWC-4.2 design artifact for the N=30-40 tensor-network baseline path. It is planning and preregistration evidence only.
TN/MPS Crossover Stage-1 Gateprimary navThis QWC-5.1 artifact admits the larger-than-16-node N=30-40 tensor-network crossover row format before any owner-gated compute run.
Topology-Aware Quantum Kernelprimary navBL-88 provides a bounded local product for asking one precise question: does an edge-aligned XY feature map retain the inductive bias of a declared coupling graph strongly enough to reproduce labels generated by that same kernel?
Issue Triage PolicycatalogThis document describes how issues and pull requests are routed, prioritised and closed in scpn-quantum-control. It exists so that a first-time contributor knows when to expect a response and what a given label means.
Pulse → UltraScale+ HLS Code Generationprimary navscpn_quantum_control.codegen.ultrascale_hls converts a quantum control pulse waveform — the output of phase/pulse_shaping.py — into a manifest-bound Vivado/Vitis HLS source artifact for AMD Xilinx Zynq UltraScale+ devices. The artifact is a
Unsuitable-scenario + anti-silent-wrong registry (BL-53)catalogThis page is the operator-facing guide for the negative-space governance product under BL-53: a versioned catalogue of scenarios that must fail closed rather than silently produce wrong gradients.
Variational Methods: Parameter-Shift Gradient Ruleprimary navscpn_quantum_control.phase.param_shift
Wirtinger (CR) Calculusprimary navscpn_quantum_control.wirtinger_calculus provides the complex-derivative surface that the registered Phase-QNode engine deliberately leaves fail-closed. The Phase-QNode differentiates real rotation angles, so its complex-derivative contr
Automated Kuramoto Witness Discoveryprimary navscpn_quantum_control.analysis.witness_discovery runs a replayable search over Kuramoto control candidates and scores each candidate with synchronisation witnesses. It is intended for small simulator sweeps and QPU pre-screening: find the pa
XY-Optimised Circuit Compilercatalogscpn_quantum_control.phase.xy_compiler

Solutions and adoption

PageSurfacePurpose
Active sensing and experimental design (BL-68)catalogscpn_quantum_control.active_sensing_product chooses the next synthetic scalar observation by expected information gain, but only after the existing BL-47 policy accepts its complete shot plan. It then runs the existing S3 analytic design pr
Adjoint reversible simulator replay product (BL-39)catalogVersioned reverse-mode adjoint-via-replay product over ambient program_ad_adjoint generation and executable step replay. Materialised local scalar demos only; refuses mid-circuit measurement / irreversible invent-green, Catalyst parity,
Adopters and Case Studiescatalogscpn-quantum-control is consumed inside the same ecosystem that builds it. These repos depend on its public API and ship changes alongside it.
Advanced witnesses product (BL-44)catalogFail-closed product surface for estimator-aware scientific diagnostics: Krylov complexity, out-of-time-order correlators (OTOC), and classical shadows, with uncertainty and provenance — beyond energy and scalar order parameters.
Application Benchmark Pluginsprimary navApplication plugins expose domain datasets through the same QPU data artifact contract used by the Kuramoto-XY pipeline. The built-in plugins cover EEG, tokamak MHD mode locking, IEEE power-grid synchronisation, and Friston-style predictive
Domain Application Honesty Kitsprimary navDomain-facing examples are useful only when their evidence boundary survives copying into notebooks, proposals, and downstream integrations. BL-63 makes that boundary executable. Each honesty kit states what the software route can demonstra
Campaign harness productisation (BL-99)catalogReusable hardware-campaign harness templates with prereg digests and BL-47 no-submit default. Dry-run probes only. Refuse invent-green live QPU submit and unattested claim promotion.
Circuit-cutting product boundary (BL-76)catalogscpn_quantum_control.circuit_cutting_product freezes and governs the existing circuit-cutting planner and partition-local simulator. It is a local, no-submit product surface; it does not claim general circuit reconstruction or large-N hardw
Cloud-native deployment boundary product (BL-101)catalogDocumented, fail-closed cloud deploy patterns for workers/batch. Dry-run manifest generation only (Kubernetes + Docker Compose). No secret leakage, no always-on QPU, no live cluster create.
QIR / CUDA-Q compiler boundary product (BL-66)catalogFirst-class external-compiler boundary register — not a marketing tick list. Status enum: supported | adapter | implementation_path | permanent_boundary.
Compose existing control/* stack product (BL-67)catalogVersioned ownership map and typed, executable adapter ports over ambient production control/* modules so co-design (BL-33) does not invent a second stack. Refuse evaluate/run without ClosedLoopExecutionPolicy. Never invent-green
Custom / registered derivatives product (BL-92 / P1)catalogVersioned safe extension surface for third-party and domain JVP/VJP rules: registration contract, public register/query helpers, fail-closed duplicates, and a documented scaled-linear example rule.
Differentiable error-mitigation product (BL-59)catalogVersioned mitigator taxonomy with differentiability classes, local ZNE and readout probes, and hard gaps for invent-green ideal-gradient restoration or live QPU mitigation.
Fault-tolerant resource product (BL-36)catalogscpn_quantum_control.fault_tolerant_resource_product turns the existing QEC resource primitives into one deterministic, digest-bound planning report for a bounded Kuramoto/XY request. It does not build another decoder or surface-code circui
Geometric quantum control product (BL-50)catalogGeometry of quantum control for synchronisation: QFI / McLachlan metric / QNG catalogue, local metric spectrum diagnostics, and regularised natural- gradient direction probes. No experimental advantage claims at criticality.
Hermetic external reproduction kit (BL-55)catalogSkeptic-facing command manifest + digest contract so public numbers can be rebuilt without trusting narrative docs alone. The core kit is local-first and must not require cloud GPUs or QPUs.
Identity and robustness control observers (BL-69)catalogscpn_quantum_control.identity_observer_product maps the existing identity robustness certificate, coherence budget, and optional CHSH witness into one fail-closed BL-33/BL-67 control decision.
KYMA / KYMA v2 mechanism benchmark product (BL-73)catalogPublic, preregistered, mechanism-only sync-learning honesty surface. Design constants come from teacher dynamics only (prereg §5) — never student held-out accuracy. Refuse post-hoc retuning and invent-green advantage without protoco
Bounded L16 director product (BL-85)primary navBL-85 turns the existing L16 indicator calculation into a policy-gated local product. It runs three frozen small-system scenarios, replays every result, maps the legacy heuristic labels into the BL-33 safety envelope, and writes digest-boun
Multi-HAL provider federation product (BL-75)catalogCapability-true federation matrix over ambient hardware/hal_* adapters and backend descriptors. Default no-submit dry-run (BL-47). Never invent-green live submit without an owner ticket; never invent online queue depth.
Neural-Operator Forecast Baselinescatalogscpn_quantum_control.neural_operator_baseline_product composes the existing classical neural-operator and observed-synchronisation forecast surfaces without adding another model or solver.
Open-system MCWF completeness product (BL-51)catalogDocumented, fail-closed open-system dynamics completeness: Lindblad and MCWF surfaces, seeded ensemble variance certificates, simulation noise-model I/O, and hard gaps for non-CP / non-Markovian / adjoint / hardware noise.
PGBO quantum geometric tensor product (BL-71)catalogMetric + Berry curvature on coupling space extracted from quantum ground states via ambient compute_pgbo_tensor. Small-system simulation only; fail- closed oscillator caps; no experimental geometry overclaim.
Phase-QNode product surface (BL-90 / P1)catalogVersioned primary quantum programming product map for Phase-QNode: public journeys, support badges, and dry-run posture. Ambient phase/qnode_* workbench modules remain experimental under BL-97 honesty (not a frozen SemVer mega-contract)
Polyglot edge Program AD (BL-74)catalogFail-closed product contract over the real bounded Rust Program-AD replay, its standalone browser WASM build, and the optional Julia boundary.
qpu_compute product surface (BL-95 / substrate)catalogFail-closed typed compute plan product between algorithms and HALs. Default posture is dry-run / no-submit; would-live and hardware_enabled plans are refused without inventing QPU spend. Composes qpu_compute_types kernels and BL-47
Quantum Sync Challenge oracle product (BL-32)catalogClaim-governed synchronisation challenge oracle façade over ambient sync witnesses, objectives, and coupling recovery. Refuse invent-green quantum advantage, unvalidated leaderboard ranks, and hardware execution without ticket.
Solutions, applications, and market valueprimary navscpn-quantum-control turns coupled-oscillator models into reproducible simulation, optimisation, hardware-readiness, and evidence workflows. Its commercial value is governed uncertainty reduction: teams can decide what is ready for a local
SSGF quantum-in-the-loop geometry gradient (BL-70)catalogscpn_quantum_control.ssgf_geometry_gradient_product governs the existing SSGF simulator path
stable_core experiment model product (BL-81 / P1)catalogVersioned public experiment model over durable Problem / Backend / Experiment / Result contracts: schema policy, JSON envelope round-trip, and digest helpers. Ambient stable_core remains the narrow durable SemVer-intent surface under BL
Stochastic estimators & policies product (BL-93 / P1)catalogVersioned finite-shot / stochastic gradient product over ambient SPSA, score-function, parameter-shift shot allocation, and confidence-policy primitives. Materialised uncertainty only; composes BL-47 no-submit honesty.
Studio executive + coverage frontier product (BL-62)catalogVersioned executive verb catalogue with BL-52 route pointers and a coverage-frontier score (honesty × answer-rate). Never invent-green full coverage while refuse rates are hidden.
Thermodynamics readiness product (BL-100)catalogHonest S9 quantum-thermodynamics readiness as a productised boundary, plus an FEP research-only inventory (BL-84 / tier C). No thermodynamic peak claim, no hardware submission, no invent-green FEP product promotion.
Fixture-driven visualisation dashboard product (BL-34)catalogVersioned static panel catalogue + fixture probes for order-parameter/energy, gradient norms, and related panel families. live_qpu=false honesty; no SaaS.
Whole-program AD product surface (BL-91 / P1)catalogVersioned frontend → IR → adjoint/replay product map for whole-program AD: public journeys, layered architecture map, support badges, and dry-run posture. Ambient whole_program_* / program_ad_* workbench modules remain experimental unde
Wirtinger + implicit differentiation product (BL-64)catalogVersioned complex Wirtinger + real implicit sensitivity product over ambient wirtinger_calculus and differentiable_implicit_sensitivity. Materialised local scalar demos only; composes BL-53 complex-without-Wirtinger refuse and BL-46 ant

Start here

PageSurfacePurpose
scpn-quantum-control — scpn-quantum-controlprimary navCI License: AGPL-3.0 Python 3.11+ Qiskit 2.2+ ![O
Installationprimary navThis page defines how to install the project with explicit capability boundaries. The same repository supports local simulation, source-level debugging, and provider-aware integration, but each use case has different dependency and reproduc
Onboardingprimary navThis page is the shortest route to understanding what scpn-quantum-control is, what it is useful for, and where the claim boundaries sit.
Quickstartprimary navAll examples run on the local AerSimulator — no IBM credentials needed.