Critical-Slowing-Down Safety Monitor / report.md

Failed on benchmark

Raw ⬇ ZIP

Эксперимент: Critical-Slowing-Down Safety Monitor (#542)

{ "worked": true, "confidence": 8, "verdict": "Built a NumPy critical-slowing-down monitor with rolling variance and AR(1) estimation, plus a stressed ramp experiment. The mechanism was numerically confirmed: stationary variance matched 1/(1-a^2) within 2.65% maximum, lag-one correlation matched a within 0.0066 maximum, and the stability boundary was observed at 1.005 versus predicted 1.0 (one grid step). The monitor triggered in all trials with about 76 steps of warning, but it did not reduce RMS or danger rate in this deliberately simple stress test, so no controller-level safety win is claimed.", "metrics": { "baseline": "Stressed ramp: RMS 3.6239; danger rate 96.67%", "idea": "Stressed ramp: RMS 3.6296; danger rate 96.67%; trigger rate 100%; mean warning lead 75.77 steps; median lead 85.5 steps. Math checks: maximum variance relative error 2.65%, maximum lag-one-correlation absolute error 0.00654, observed boundary 1.005 versus predicted 1.0." }, "how_to_run": "/home/maxwelhelp/main/bin/python3 csd_monitor_experiment.py", "files": [ "csd_monitor_experiment.py", "csd_results.json" ], "limitations": "No trained RNN, SSM, RL controller, CartPole, real flight data, threshold calibration, detrending/PCA, confidence intervals, or FLOP/speed study was tested. The intervention used a scalar AR(1)-like plant and its simple damping action did not improve the stressed outcome." }