SCPN-Quantum-Control

A quantum simulation framework that uses IBM's superconducting processors to test how networks of coupled oscillators synchronise. Classical Kuramoto theory in, noisy hardware out, and the gap in between is where the physics lives.

31Domain Families
699Python Modules
177Rust Bindings
1,155Test Files
106Notebooks
342ibm_kingston circuits
+17.5%Peak DLA asymmetry
1,665×Peak Rust speedup

Hardware Evidence — amended July 2026

Parity-sector and excitation-number correlated leakage asymmetry

On ibm_kingston, a 342-circuit, four-qubit campaign across eight Trotter depths recorded a +17.48% peak contrast at depth 6. Follow-up popcount controls found substantial same-popcount within-sector variation, so the current evidence does not support attributing the contrast to DLA parity alone.

The promoted statement is therefore conservative: the committed raw counts show a backend-sensitive, parity-sector and excitation-number correlated leakage asymmetry. Result packs, job identifiers, analysis scripts, and dated amendments remain public; mechanism claims follow the hardware-status ledger.

Publications — Zenodo Preprints

Seven open-access preprints (25 May 2026) document the hardware campaigns, methods, and software behind this project. Each is archived on Zenodo with a citable DOI and carries the claim boundaries stated in its own abstract. These are preprints under the project's evidence discipline — not peer-reviewed journal articles.

Install

# Core framework
pip install scpn-quantum-control

# With IBM hardware + Rust acceleration + GPU
pip install scpn-quantum-control[ibm,rust,gpu]

# Everything
pip install scpn-quantum-control[all]

Architecture Pipeline

CLASSICAL SIDE QUANTUM SIDE Knm matrix SCPN layer 12 XY Hamiltonian KnmCompiler Trotter / VQE Qiskit circuits DynQ + ICI pulses qubit placement + control IBM Heron r2 ibm_kingston / ibm_fez Bitstring counts SamplerV2 GUESS mitigation symmetry-guided ZNE DLA parity analysis Welch + Fisher Knm coupling matrix → XY Hamiltonian → Trotter compilation → DynQ placement → Heron r2 → bitstring counts → GUESS mitigation → DLA parity analysis Every box in this diagram is a clickable page.

Known Limitations

What this project is, and what it isn't

  • Stable release v1.1.0, active development. Stable public contracts are versioned, while research and hardware claim surfaces continue to evolve through dated amendments. Pin exact artefacts for reproducible results.
  • Coherence wall at depth 250–400 two-qubit gates. Beyond this depth, decoherence dominates on Heron r2 and error mitigation reduces but does not eliminate systematic bias. This is a hard NISQ-era constraint, not specific to our protocol.
  • Hardware access is private. The IBM Quantum Platform account is personal (Open Plan). Reviewers cannot directly re-execute campaigns without separate access; all raw counts, job identifiers, and analysis scripts are committed so the results can be checked without re-running the hardware.
  • Backend and state-preparation confounds remain. Follow-up controls narrow the promoted wording to parity-sector and excitation-number correlated leakage; DLA parity alone is not established as the mechanism.
  • Capability inventory. The active v1.1.0 tree contains 697 Python source modules, 1,378 public classes, 177 Rust bindings, 1,371 Python test files, 841 public API exports, 109 notebooks, 38 domain families, and 37 examples. These are static counts, not performance evidence.
  • Open source, public repository. Source and evidence are public under AGPL-3.0 with a commercial licensing route for proprietary deployment.

Explore

Phase Synchronisation →
Drive the Kuramoto transition live — the library's exact mean-field RK4 kernel, running in your browser.
Network Topology →
The same oscillators, different wiring — watch topology decide synchronisation with the exact networked RK4 kernel.
Trotterisation →
The cost of chopping time — the product-formula error made exact on one qubit, converging as you add steps.
Geometric (Berry) Phase →
The phase a state keeps of its path — minus half the enclosed solid angle, computed live two independent ways.
ICI Pulse Shaping →
Pump and Stokes drives trading power along a mixing angle — the ICI sequence, live from the library's exact formulas.
DLA Parity Sectors →
The XY algebra splits as su(2^{n-1}) ⊕ su(2^{n-1}) — build the even/odd sectors exactly and see the symmetric-noise leakage baseline.
BKT Transition →
Order without an order parameter — vortex unbinding, algebraic-to-exponential correlations and a diverging length, from the library's exact BKT universals.
Quantum Error Correction →
The threshold, exactly — below it redundancy suppresses the logical error, above it makes things worse, plus the parity-protected subspace angle.
Quantum Sensing →
An atom-sized magnetometer — the NV-centre ground-state Hamiltonian diagonalised live, the resonances a field moves, through the 102 mT anti-crossing.
Quantum Neuromorphic →
A spiking neuron on a qubit — the membrane integrates, the voltage becomes a rotation, the spike is a measurement. Live quantum LIF.
Quantum Reservoir Computing →
Leave the dynamics fixed, train only a linear readout — the echo-state baseline on a memory task, with the edge-of-chaos sweet spot.
Quantum Thermodynamics →
Work is a distribution — the Jarzynski equality recovers the free energy exactly, and dissipation is never negative on average. Live.
Hypergeometric Pulses →
One (α,β) family holding Allen–Eberly, STIRAP and Demkov–Kunike — built live from the library's exact Gauss ₂F₁ formula.
Phase 1 Results →
Full DLA parity dataset, interactive Plotly plot, reproducibility pointer, BibTeX.
The Science →
Plain-language primer on SCPN, Kuramoto-XY, and why DLA parity matters.
GUESS Mitigation →
Shot-budget-free symmetry-guided ZNE for the XY Hamiltonian.
DLA Parity Theorem →
Derivation of the $\mathfrak{su}(2^{n-1}) \oplus \mathfrak{su}(2^{n-1})$ decomposition.
Pulse Shaping →
ICI (1,665× Rust) and (α, β)-hypergeometric (44× Rust) control.
Algorithms →
VQE, ADAPT-VQE, Trotter, VarQITE, QSVT, Floquet DTC, Lindblad, MCWF.
697 Modules →
38 domain families and 841 public API exports in the active v1.1.0 capability inventory.
Benchmarks →
Rust vs Python performance, Phase 1 DLA parity table, coherence wall analysis.
Hardware Validation →
February 2026 ibm_fez campaign: Bell tests, QKD, VQE ground state, qubit scaling.
Timeline →
Past milestones, current status, planned Phase 2, speculative Phase 4.
Reproducibility →
Per-commit pinning, pinned deps, IBM job IDs, rerun protocol.
Access & Citation →
PyPI install, source access for reviewers, pricing, BibTeX.
Applications →
Where coupled-oscillator quantum control is aimed — honestly scoped use cases.
Notebooks →
Runnable notebooks — from quickstart to the DLA-parity reproduction.
Learning Path →
One route through the portal in eight levels — what to learn, what to try, and a checkpoint at each.
Glossary →
Every term in one searchable place — Kuramoto, XY, DLA parity, Trotter, GUESS, Berry, BKT, STIRAP.
Papers →
Downloadable, Zenodo-archived manuscripts — the synchronisation study, the DLA-parity observation, Rust/VQE methods, a hardware falsification, and a negative control experiment.