Эксперимент: Critical-Gain Covariance Controller (#607)

{ "worked": false, "confidence": 10, "verdict": "Built the exact covariance Jacobian diagnostic, fixed-point sweep, and a small CUDA RNN comparison with the proposed gain/covariance penalty. The stated Delta eigenvalue formula failed with maximum absolute error 3.2; symbolic and numerical checks instead confirmed eigenvalues {-2, 2(kE+kI-1), kE+kI-2} to 4.0e-12, with the stability boundary at kE+kI=1. The measured decay-rate scaling matched the corrected prediction closely and fixed-point log-log scaling was -0.945 versus -1.0, but the proposed controller did not reduce hidden covariance and only modestly improved long loss (0.7053 vs 0.7456).", "metrics": { "baseline": "long_loss=0.745608; long_trace=15.7703; long_delta=-0.8958", "idea": "long_loss=0.705277; long_trace=15.7369; long_delta=-0.8989; stated-eigenvalue max error=3.2; corrected-eigenvalue max error=4.04e-12; corrected decay scaling measured/predicted near 1.0; fixed-point slope=-0.945/-1 predicted" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py", "files": [ "experiment.py", "results.json" ], "limitations": "Only a tiny synthetic next-symbol RNN was tested, not sequential MNIST, language modeling, GRU/SSM architectures, spectral normalization, or multi-seed statistical comparisons. The supplied paper formula was not reconciled against a broader model variant because the displayed J directly contradicts the stated Delta eigenvalues." }