Эксперимент: Information-Budgeted Reverse-Dynamics Controller (#419)

{ "worked": true, "confidence": 8, "verdict": "Built a reproducible linear-Gaussian MVP in experiment.py. The three quantitative predictions manifested: the Gaussian information estimator matched I=0.5 log(1+V/r) within 0.248%, reverse-kernel KL was zero while passive-to-reverse KL increased with contraction radius, and increasing beta produced a monotone information frontier. The control effect was real but weak: beta=8 reduced information by 0.089% and increased task cost by 0.095%, so this supports the mechanism but not a meaningful performance win.", "metrics": { "baseline": "beta=0: task cost 0.304178, information 0.308190 nats/step, k=-0.33738", "idea": "beta=8: task cost 0.304466, information 0.307916 nats/step, k=-0.35541; information decreased and task cost increased monotonically across the beta sweep", "information_math": "Maximum relative error between predicted and measured information over five noise levels: 0.248%", "reverse_kernel_math": "Reverse realization KL was 0 for radii 0.2, 0.4, 0.6, 0.82, and 0.92; passive-to-reverse KL increased from 0.02763 to 3.66567 nats" }, "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py", "files": [ "experiment.py" ], "limitations": "The MVP uses analytically optimized linear-Gaussian policies rather than a trained GRU/RNN, has no controller memory H, no learned variational marginal q_phi(z|h), and does not test neural-controller optimization, target-distribution mismatch, runtime, or high-dimensional plants." }