Free energies of Boltzmann Machines: self-averaging, annealed and replica symmetric approximations in the thermodynamic limit
Elena Agliari, Adriano Barra, Brunello Tirozzi

TL;DR
This paper investigates the free energies of Restricted Boltzmann Machines (RBMs) using advanced statistical mechanics methods, deriving self-averaging properties, annealed bounds, and replica symmetric solutions for different weight types, and analyzing fluctuations related to glassy behavior.
Contribution
It extends analytical techniques to study RBMs' free energies, providing rigorous results for Boolean and Gaussian weights, including self-averaging, annealed bounds, and replica symmetry analysis.
Findings
Proved self-averaging of free energy in the thermodynamic limit.
Derived the annealed bound for RBMs with digital couplings.
Identified the replica symmetry critical line and analyzed overlap fluctuations.
Abstract
Restricted Boltzmann machines (RBMs) constitute one of the main models for machine statistical inference and they are widely employed in Artificial Intelligence as powerful tools for (deep) learning. However, in contrast with countless remarkable practical successes, their mathematical formalization has been largely elusive: from a statistical-mechanics perspective these systems display the same (random) Gibbs measure of bi-partite spin-glasses, whose rigorous treatment is notoriously difficult. In this work, beyond providing a brief review on RBMs from both the learning and the retrieval perspectives, we aim to contribute to their analytical investigation, by considering two distinct realizations of their weights (i.e., Boolean and Gaussian) and studying the properties of their related free energies. More precisely, focusing on a RBM characterized by digital couplings, we first extend…
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.
