An Information-Theoretic Route to Isoperimetric Inequalities via Heat Flow and Entropy Dissipation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Sangha, Amandip
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914164324171776
author Sangha, Amandip
author_facet Sangha, Amandip
contents We develop an information-theoretic approach to isoperimetric inequalities based on entropy dissipation under heat flow. By viewing diffusion as a noisy information channel, we measure how mutual information about set membership decays over time. This decay rate is shown to be determined by the boundary measure of the set, leading to a new proof of the Euclidean isoperimetric inequality with its sharp constant. The method extends to Riemannian manifolds satisfying curvature-dimension conditions, yielding Levy-Gromov and Gaussian isoperimetric results within a single analytic principle. Quantitative and stability bounds follow from refined entropy inequalities linking information loss to geometric rigidity. The approach connects geometric analysis and information theory, revealing how entropy dissipation encodes the geometry of diffusion and boundary.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15356
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle An Information-Theoretic Route to Isoperimetric Inequalities via Heat Flow and Entropy Dissipation
Sangha, Amandip
Differential Geometry
Information Theory
Analysis of PDEs
Metric Geometry
Probability
We develop an information-theoretic approach to isoperimetric inequalities based on entropy dissipation under heat flow. By viewing diffusion as a noisy information channel, we measure how mutual information about set membership decays over time. This decay rate is shown to be determined by the boundary measure of the set, leading to a new proof of the Euclidean isoperimetric inequality with its sharp constant. The method extends to Riemannian manifolds satisfying curvature-dimension conditions, yielding Levy-Gromov and Gaussian isoperimetric results within a single analytic principle. Quantitative and stability bounds follow from refined entropy inequalities linking information loss to geometric rigidity. The approach connects geometric analysis and information theory, revealing how entropy dissipation encodes the geometry of diffusion and boundary.
title An Information-Theoretic Route to Isoperimetric Inequalities via Heat Flow and Entropy Dissipation
topic Differential Geometry
Information Theory
Analysis of PDEs
Metric Geometry
Probability
url https://arxiv.org/abs/2511.15356