Skip to content

README.md

Latest commit

 

History

History
78 lines (51 loc) · 3.3 KB

README.md

File metadata and controls

78 lines (51 loc) · 3.3 KB

CI Coverage DOI Python Code style: ruff

QISA Consensus Engine

Deterministic, auditable multi-perspective consensus with explicit fixpoint semantics and tamper-evident trace hashing.

QISA provides a reference-grade core for decision systems where reproducibility, auditability, and determinism are first-class constraints.

This repository intentionally avoids stochasticity and external services in order to support verifiable execution and post-hoc inspection.

Key properties

  • Deterministic convergence
    Fixpoint iteration converges (or fails fast) under bounded steps.

  • Idempotence
    Re-running from a converged state yields the same final state and trace hash.

  • Tamper-evident traces
    Each execution step is hash-chained; any post-hoc mutation is detectable.

  • Reproducible benchmarks
    Benchmarks emit pinned JSON artifacts plus a generated comparison table.

What problem does QISA solve?

QISA (Quantum-Inspired System of AI Consensus) addresses a core problem in automated decision systems:
how to produce reproducible, auditable, and deterministic decisions when multiple perspectives disagree.

In five clear points

  1. Deterministic consensus (non-stochastic) — same inputs, same outputs.
  2. Verifiable audit trail — hash-chained traces detect tampering.
  3. Fixpoint convergence — stability via idempotence prevents loops.
  4. Comparable baselines — reproducible benchmark harness + pinned results.
  5. No external dependencies — not an LLM, no external services required.

QISA does not aim to be creative. It aims to be correct, verifiable, and repeatable.

Quality gate: CI enforces test coverage (pytest-cov) with a minimum threshold.

Installation

python -m venv .venv
source .venv/bin/activate  # Windows: .\.venv\Scripts\Activate.ps1
python -m pip install -U pip
python -m pip install -e ".[dev]"

<!-- QISA_META_START -->

## Tools (Trace export + external verification)
- Paper-grade instructions: 	ools/README.md
- Quick demo:
  - Host: python tools\export_trace_demo.py then python tools\verify_trace_json.py tools\_artifacts\trace_demo.json
  - Docker: docker run --rm -v \C:\repos\qisa-consensus-engine:/app qisa-consensus-engine:dev python tools/export_trace_demo.py

## Benchmarks (adversarial)
- Latest results (generated): �enchmarks\_results\results.md
- Re-run: python .\benchmarks\bench_adversarial.py

<!-- QISA_META_END -->

---

## Kemrai Linkage

This repository is part of the technical and scientific foundation behind **Kemrai** (https://www.kemrai.com).  
It exists as public evidence of rigorous engineering (determinism, reproducibility, traceability) and research-grade methodology.