Sobolev Gradient Ascent for Optimal Transport: Barycenter Optimization and Convergence Analysis

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kim, Kaheon, Zhou, Bohan, Zhu, Changbo, Chen, Xiaohui
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915942516129792
author Kim, Kaheon
Zhou, Bohan
Zhu, Changbo
Chen, Xiaohui
author_facet Kim, Kaheon
Zhou, Bohan
Zhu, Changbo
Chen, Xiaohui
contents This paper introduces a new constraint-free concave dual formulation for the Wasserstein barycenter. Tailoring the vanilla dual gradient ascent algorithm to the Sobolev geometry, we derive a scalable Sobolev gradient ascent (SGA) algorithm to compute the barycenter for input distributions discretized over a regular grid. Despite the algorithmic simplicity, we provide a global convergence analysis that achieves the same rate as the classical subgradient descent methods for minimizing nonsmooth convex functions in the Euclidean space. A central feature of our SGA algorithm is that the computationally expensive $c$-concavity projection operator enforced on the Kantorovich dual potentials is unnecessary to guarantee convergence, leading to significant algorithmic and theoretical simplifications over all existing primal and dual methods for computing the exact barycenter. Our numerical experiments demonstrate the superior empirical performance of SGA over the existing optimal transport barycenter solvers.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13660
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sobolev Gradient Ascent for Optimal Transport: Barycenter Optimization and Convergence Analysis
Kim, Kaheon
Zhou, Bohan
Zhu, Changbo
Chen, Xiaohui
Optimization and Control
Machine Learning
This paper introduces a new constraint-free concave dual formulation for the Wasserstein barycenter. Tailoring the vanilla dual gradient ascent algorithm to the Sobolev geometry, we derive a scalable Sobolev gradient ascent (SGA) algorithm to compute the barycenter for input distributions discretized over a regular grid. Despite the algorithmic simplicity, we provide a global convergence analysis that achieves the same rate as the classical subgradient descent methods for minimizing nonsmooth convex functions in the Euclidean space. A central feature of our SGA algorithm is that the computationally expensive $c$-concavity projection operator enforced on the Kantorovich dual potentials is unnecessary to guarantee convergence, leading to significant algorithmic and theoretical simplifications over all existing primal and dual methods for computing the exact barycenter. Our numerical experiments demonstrate the superior empirical performance of SGA over the existing optimal transport barycenter solvers.
title Sobolev Gradient Ascent for Optimal Transport: Barycenter Optimization and Convergence Analysis
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2505.13660