Эксперимент: Lunar Color-Connectivity Regularizer (#1206)

{ "worked": true, "confidence": 7, "verdict": "Built a two-color lunar EMST cost verifier using exact brute-force minimum enclosing circles, plus a small PyTorch classification experiment with a differentiable nearest cross-color surrogate. The math check produced nonnegative costs for 20 random configurations, and at identical 100% classification and 3-NN accuracy the regularizer reduced cross-color nearest distance from 0.426 to 0.156 and normalized lunar cost from 0.750 to 0.687. This is a promising connectivity signal, but not an accuracy win; the experiment is too small to establish generalization gains.", "metrics": { "baseline": "CE: accuracy=1.000, kNN3=1.000, cross-color nearest=0.4260, lunar cost=4.1656, normalized lunar cost=0.7505", "idea": "lambda=0.5: accuracy=1.000, kNN3=1.000, cross-color nearest=0.1560, lunar cost=3.5730, normalized=0.6870; lambda=2: accuracy=1.000, kNN3=1.000, cross-color nearest=0.1159, lunar cost=2.9533, normalized=0.6893" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 lunar_experiment.py", "files": [ "lunar_experiment.py", "results.json" ], "limitations": "No CIFAR-10 or ResNet experiment, no multi-seed statistics, and no full differentiable hierarchy implementation. The training penalty is a nearest cross-color approximation rather than the exact lunar cost; the toy data are linearly structured and classification is already saturated." }