Эксперимент: Cycle-Aware Heavy-Ball Safeguard (#1194)

{ "worked": false, "confidence": 9, "verdict": "Built a projected cycle-aware heavy-ball optimizer diagnostic with exact 2x2 spectral-radius computation and a quadratic stress test. The matrix recurrence matched exactly, and reducing learning rate/momentum lowered the measured spectral radius from approximately 1.000 to 0.815, confirming the local stability diagnostic. However, the safeguard produced much worse divergence than fixed momentum in the tested ill-conditioned quadratic, and the specified lag-2 correlation is -0.5 for an ideal period-three orbit, so it does not detect the paper’s explicit cycle pattern.", "metrics": { "baseline": "Quadratic: final loss 5.596e3, best loss 338.1, loss at step 100 5.533e3, triggers 0", "idea": "Quadratic: final loss 4.158e65, best loss 338.1, loss at step 100 1.014e17, triggers 1; math matrix error 0, rho 1.000 before and 0.815 after reduction; ideal period-3 lag-2 correlation -0.5" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 cycle_safeguard.py", "files": [ "cycle_safeguard.py" ], "limitations": "Only a small deterministic numpy quadratic experiment was tested; no MLP, Transformer, CIFAR-10/TinyStories task, AdamW integration, Hessian-vector-product curvature estimate, multi-seed study, FLOP comparison, or smooth piecewise-quadratic attracting cycle was evaluated. The recovery and threshold schedule remains empirical and the tested reduction was insufficient to stabilize this quadratic." }