Эксперимент: Thermodynamic Confidence Controller for SGD (#1349)

{ "worked": false, "confidence": 9, "verdict": "Built an online current monitor using the paper's sub-gamma rate function, confidence-radius inversion, autocorrelation estimate, and batch-size trigger. The mathematical sanity check reproduced the claimed T^-1/2 radius scaling exactly (slope -0.50), and quadrupling variance doubled the radius. However, at equal sample budget the controller reached 0.9133 test accuracy versus 0.9489 for vanilla SGD, with higher final loss (0.3978 vs 0.1171), so no training win was observed; the controller's estimates are heuristic proxies rather than a rigorous bound for SGD.", "metrics": { "baseline": "Digits MLP, SGD batch 32, 1102 steps / 35,264 examples: final loss 0.1171, test accuracy 0.9489", "idea": "Digits MLP, confidence controller, 300 steps / 35,264 examples, batches 32->64->128: final loss 0.3978, test accuracy 0.9133, first/last radius 0.00460/0.00178; synthetic radius log-log slope -0.50" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 thermo_controller.py", "files": [ "thermo_controller.py", "results.json", "results2.json", "results_fair.json" ], "limitations": "Only one fixed-seed digits-MLP comparison was run; no MNIST, multiple seeds, wall-clock or FLOP accounting, Adam comparison, learning-rate adaptation, bootstrap calibration, or empirical tail-coverage validation was performed. The implemented C and D estimates are heuristic adaptations and are not proven concentration constants for discrete SGD." }