Эксперимент: Action-calibrated cycle-hopping RNN (#986)

{ "worked": true, "confidence": 8, "verdict": "Built and numerically verified a two-cycle stochastic continuous-time latent system with action estimation, discrete minimum-action optimization, Floquet stability checks, and transition-rate sweeps. The action estimate agreed with the finite-endpoint integral within 0.07%, transverse Floquet multipliers were 0.0432 for both cycles, and the rare-noise Arrhenius slope was -0.01481 versus the predicted -0.015625 (5.2% error, R²=0.999). At high noise the fitted slope shifted to -0.0547, confirming breakdown of the rare-event law; this validates the mechanism in the toy setting, but not neural sequence prediction.", "metrics": { "baseline": "No neural baseline was run; this experiment tests the mechanism directly. Predicted action barrier=0.015625, rare-regime fitted slope=-0.01481, R²=0.9992, Floquet multipliers=0.0432, predicted noise crossover D≈0.0199.", "idea": "Discrete action estimate=0.01549 versus finite-endpoint integral=0.01548 (0.07% relative error); rare-regime rate slope agrees within 5.2%; high-noise slope=-0.0547 versus rare slope=-0.0148, demonstrating non-Arrhenius crossover." }, "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py", "files": [ "experiment.py", "results.json" ], "limitations": "This is a low-dimensional radial toy system rather than a trained GRU, Neural ODE, or world model. It does not test long-horizon multimodal prediction, sample efficiency, learned action calibration, prefactor estimation, anisotropic noise, or comparison against a standard RNN. The transition estimator uses finite-step Euler simulation and a separatrix threshold, and the action optimization uses fixed duration and endpoints." }