Neural solver for Wasserstein Geodesics and optimal transport dynamics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Hailiang, Chen, Yan-Han
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908852433190912
author Liu, Hailiang
Chen, Yan-Han
author_facet Liu, Hailiang
Chen, Yan-Han
contents In recent years, the machine learning community has increasingly embraced the optimal transport (OT) framework for modeling distributional relationships. In this work, we introduce a sample-based neural solver for computing the Wasserstein geodesic between a source and target distribution, along with the associated velocity field. Building on the dynamical formulation of the optimal transport (OT) problem, we recast the constrained optimization as a minimax problem, using deep neural networks to approximate the relevant functions. This approach not only provides the Wasserstein geodesic but also recovers the OT map, enabling direct sampling from the target distribution. By estimating the OT map, we obtain velocity estimates along particle trajectories, which in turn allow us to learn the full velocity field. The framework is flexible and readily extends to general cost functions, including the commonly used quadratic cost. We demonstrate the effectiveness of our method through experiments on both synthetic and real datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2602_22003
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Neural solver for Wasserstein Geodesics and optimal transport dynamics
Liu, Hailiang
Chen, Yan-Han
Machine Learning
Optimization and Control
93E20 (Primary), 49Q22 (Secondary)
In recent years, the machine learning community has increasingly embraced the optimal transport (OT) framework for modeling distributional relationships. In this work, we introduce a sample-based neural solver for computing the Wasserstein geodesic between a source and target distribution, along with the associated velocity field. Building on the dynamical formulation of the optimal transport (OT) problem, we recast the constrained optimization as a minimax problem, using deep neural networks to approximate the relevant functions. This approach not only provides the Wasserstein geodesic but also recovers the OT map, enabling direct sampling from the target distribution. By estimating the OT map, we obtain velocity estimates along particle trajectories, which in turn allow us to learn the full velocity field. The framework is flexible and readily extends to general cost functions, including the commonly used quadratic cost. We demonstrate the effectiveness of our method through experiments on both synthetic and real datasets.
title Neural solver for Wasserstein Geodesics and optimal transport dynamics
topic Machine Learning
Optimization and Control
93E20 (Primary), 49Q22 (Secondary)
url https://arxiv.org/abs/2602.22003