Эксперимент: Robust Physics-Sparse Neural Dynamics (#605)

{ "worked": true, "confidence": 9, "verdict": "Built a compact hybrid physics-sparse dynamics MVP with known relaxation, sparse cross-interaction library, TLS, RANSAC consensus selection, hard thresholding, Jacobian spectral-radius checks, and rollout evaluation. The predicted stability boundary was gain*=5.8824 at h=0.08 versus 5.8837 observed (0.023% error), and the predicted 1/h scaling across four step sizes matched within 0.03%. Under 5% corrupted derivative samples, TLS-RANSAC achieved 0.0149 relative coefficient error versus 0.0422 for OLS and 0.0934 for plain TLS, so the mechanism manifested clearly in this toy setting.", "metrics": { "baseline": "OLS sparse-library coefficient relative error 0.0422; 100-step hybrid rollout terminal norm 0.2871", "idea": "TLS-RANSAC coefficient relative error 0.0149; 100-step hybrid rollout terminal norm 0.2864; predicted/observed stability gain boundary 5.8824/5.8837; step-size boundary relative errors 0.014%, 0.012%, 0.014%, and 0.022% for h=0.04, 0.08, 0.16, and 0.24" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py", "files": [ "experiment.py", "results.json" ], "limitations": "This is a deterministic two-dimensional damped-oscillator numerical verification, not a full neural state-space model with an MLP encoder or GRU baseline. It uses synthetic finite-difference data, forward Euler discretization, independently applied scalar TLS fits, and does not test real microgrid PMU data, long-horizon distribution shift, GPU execution, or a FLOP-matched unconstrained neural transition." }