PreprintarXiv licencearXiv
Sharp Stability of Brenier Potentials and its Link to Poincaré Inequalities
Abstract
We prove a new functional inequality for uniformly log-concave measures, stronger than the Poincaré inequality. This new inequality bounds the variance of differences of convex functions and is motivated by the stability theory of optimal transport: for a fixed uniformly log-concave source measure, it provides a sharp stability bound for Brenier potentials with respect to variations of the target measure. Applying this inequality to Gaussian measures, we recover the fact that the Bures-Wasserstein distance between two positive semidefinite matrices is equivalent to the Frobenius distance between their square roots. Our proofs hinge upon linearization arguments for the Monge-Ampère equation governing the quadratic optimal transport problem.
Related
- Caffarelli-Kohn-Nirenberg and Weighted Gaussian Poincaré Inequalities: a complete characterization of sharp L2 stability and Lp extensions
- Sharp L2-stability and gradient stability hierarchies of the L2-Poincaré inequalities on Euclidean balls and Gaussian Poincaré inequality
- The stability of Sharp Poincaré and Brascamp--Lieb inequalities with full monomial weights
- Sharp potential and Brenier map W2-stability for log-concave sources
- Sharp Nonlocal Poincaré Inequalities and Liminf Lower Bounds for Sobolev and BV Energies on Metric Measure Spaces
BibTeX
@misc{letrouit2026sharp,
title = {{Sharp Stability of Brenier Potentials and its Link to Poincaré Inequalities}},
author = {Cyril Letrouit and Tudor Manole},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.37659},
eprint = {2610.37659},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.37659},
}
RIS
TY - UNPB
TI - Sharp Stability of Brenier Potentials and its Link to Poincaré Inequalities
AU - Cyril Letrouit
AU - Tudor Manole
PY - 2026
DA - 2026/10/27
PB - arXiv
DO - 10.48550/arxiv.2610.37659
UR - https://arxiv.org/abs/2610.37659
AB - We prove a new functional inequality for uniformly log-concave measures, stronger than the Poincaré inequality. This new inequality bounds the variance of differences of convex functions and is motivated by the stability theory of optimal transport: for a fixed uniformly log-concave source measure, it provides a sharp stability bound for Brenier potentials with respect to variations of the target measure. Applying this inequality to Gaussian measures, we recover the fact that the Bures-Wasserstein distance between two positive semidefinite matrices is equivalent to the Frobenius distance between their square roots. Our proofs hinge upon linearization arguments for the Monge-Ampère equation governing the quadratic optimal transport problem.
ER -