Local Max-Entropy and Free Energy Principles Solved by Belief Propagation
Olivier Peltre

TL;DR
This paper demonstrates that generalized belief propagation efficiently solves local variational principles related to free energy and entropy in statistical systems with local interactions, extending previous theoretical correspondences.
Contribution
It shows GBP converges to critical points of Bethe-Kikuchi approximations, linking local variational principles with global thermodynamic functions.
Findings
GBP solves local variational principles for free energy and entropy.
Convergence to critical points of Bethe-Kikuchi approximations.
Extension of Yedidia et al.'s correspondence to local variational principles.
Abstract
A statistical system is classically defined on a set of microstates by a global energy function , yielding Gibbs probability measures (softmins) for every inverse temperature . Gibbs states are simultaneously characterized by free energy principles and the max-entropy principle, with dual constraints on inverse temperature and mean energy respectively. The Legendre transform relates these diverse variational principles which are unfortunately not tractable in high dimension. The global energy is generally given as a sum of local short-range interactions indexed by bounded subregions , and this local structure can be used to design good approximation…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsStatistical Mechanics and Entropy · Advanced Thermodynamics and Statistical Mechanics · Quantum many-body systems
