Эксперимент: Anchored Whitening Layer (#923)

{ "worked": true, "confidence": 9, "verdict": "Built an anchored whitening MVP using T=R^{-1/2}Q with identity-preserving Q=I, plus deterministic mathematical sweeps and a toy classification comparison. Whitening residuals stayed between 0 and 1.2e-14, measured fidelity matched the analytic curve within 1.5e-15, and threshold feasibility agreed with prediction in all 10 tested cases. The toy classifier reduced off-diagonal covariance substantially without changing accuracy, so no accuracy win was observed.", "metrics": { "baseline": "Per-channel standardization: accuracy=0.889714; off-diagonal covariance Frobenius norm=3.31448", "idea": "Q=I ZCA/anchored whitening: accuracy=0.889714; off-diagonal covariance Frobenius norm=1.899e-15; maximum covariance error=1.055e-15; minimum sample fidelity=0.8523", "math_predictions": "Exact decorrelation error was 0 to 1.18e-14; fidelity-curve absolute error was 0 to 1.44e-15 over correlations {0, 0.1, 0.3, 0.6, 0.9}; threshold transition agreed in 10/10 cases" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py > results.json", "files": [ "experiment.py", "results.json" ], "limitations": "The MVP tests the exact whitening family and identity anchor, not a learned orthogonal Q optimized with a penalty or Cayley update. The empirical comparison is a small synthetic logistic-regression task, not CIFAR/ResNet, and does not test EMA covariance, training dynamics, gradient noise, FLOPs, or BatchNorm/LayerNorm neural-network baselines." }