The Wasserstein gradient flow of the Sinkhorn divergence between Gaussian distributions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hardion, Mathis, Lacombe, Théo
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915805476683776
author Hardion, Mathis
Lacombe, Théo
author_facet Hardion, Mathis
Lacombe, Théo
contents We study the Wasserstein gradient flow of the Sinkhorn divergence when both the source and the target are Gaussian distributions. We prove the existence of a flow that stays in the class of Gaussian distributions, and is unique in the larger class of measures with strongly-concave and smooth log-densities. We prove that the flow globally converges toward the target measure when the source's covariance matrix is not singular, and provide counter-examples to global convergence when it is, giving a first answer to an open question raised in [Carlier et al. 2024, \S4.2]. When the covariance matrix of the source distribution commutes with that of the target, we derive more quantitative results that showcase exponential convergence toward the target when the source and the target share their support, but dropping to linear rates (O(t^{-1})) if the target is concentrated on a strict subspace of the source's support.
format Preprint
id arxiv_https___arxiv_org_abs_2602_10726
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Wasserstein gradient flow of the Sinkhorn divergence between Gaussian distributions
Hardion, Mathis
Lacombe, Théo
Analysis of PDEs
Optimization and Control
We study the Wasserstein gradient flow of the Sinkhorn divergence when both the source and the target are Gaussian distributions. We prove the existence of a flow that stays in the class of Gaussian distributions, and is unique in the larger class of measures with strongly-concave and smooth log-densities. We prove that the flow globally converges toward the target measure when the source's covariance matrix is not singular, and provide counter-examples to global convergence when it is, giving a first answer to an open question raised in [Carlier et al. 2024, \S4.2]. When the covariance matrix of the source distribution commutes with that of the target, we derive more quantitative results that showcase exponential convergence toward the target when the source and the target share their support, but dropping to linear rates (O(t^{-1})) if the target is concentrated on a strict subspace of the source's support.
title The Wasserstein gradient flow of the Sinkhorn divergence between Gaussian distributions
topic Analysis of PDEs
Optimization and Control
url https://arxiv.org/abs/2602.10726