arXiv · 2606.16399
Non-Equilibrium Model Selection via Finite-Time Thermodynamics
Abstract
Information criteria such as WAIC and WBIC extend model selection to singular learning machines, but they are usually derived for equilibrium posteriors. We formulate a finite-time analogue of WBIC by replacing the equilibrium posterior with an effective ensemble generated by learning dynamics under a resource constraint. When this ensemble admits an analytic effective potential, singular learning theory yields a resource-dependent real log canonical threshold. The resulting estimator gives a computable thermodynamic contribution to time-bounded MDL and identifies the finite-time singular complexity relevant to the structural information measured by epiplexity.
Explore related subjects
Keep this discovery
Masayuki Ohzeki. 2026-06-15. Non-Equilibrium Model Selection via Finite-Time Thermodynamics. https://arxiv.org/abs/2606.16399
Cite the original work for its findings. Save a collection to share your selection of sources.