Erasure entropies and Gibbs measures
Aernout van Enter, Evgeny Verbitskiy

TL;DR
This paper explores the relationship between erasure entropies and Gibbs measures, providing formulas for erasure entropy in Gibbs measures and exact solutions for certain 2D Ising models, advancing understanding in statistical mechanics.
Contribution
It establishes a detailed connection between erasure entropies and Gibbs measures, including explicit formulas and exact solutions for specific models.
Findings
Derived a formula for erasure entropy of Gibbs measures.
Provided an exact solution for erasure entropy in 2D Ising models.
Clarified the relation between erasure entropies and statistical mechanics concepts.
Abstract
Recently Verdu and Weissman introduced erasure entropies, which are meant to measure the information carried by one or more symbols given all of the remaining symbols in the realization of the random process or field. A natural relation to Gibbs measures has also been observed. In his short note we study this relation further, review a few earlier contributions from statistical mechanics, and provide the formula for the erasure entropy of a Gibbs measure in terms of the corresponding potentia. For some 2-dimensonal Ising models, for which Verdu and Weissman suggested a numerical procedure, we show how to obtain an exact formula for the erasure entropy. l
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 · Theoretical and Computational Physics · Neural Networks and Applications
