English

Sliced optimal transport: is it a suitable replacement?

Metric Geometry 2024-05-29 v3 Optimization and Control Probability

Abstract

We introduce a one-parameter family of metrics on the space of Borel probability measures on Euclidean space with finite ppth moment for 1p<1\leq p <\infty, called the sliced Monge–Kantorovich metrics\textit{sliced Monge--Kantorovich metrics}, which include the sliced Wasserstein and max-sliced Wasserstein metrics. We then show that these are complete, separable metric spaces that are topologically equivalent to the classical Monge--Kantorovich metrics and these metrics have a dual representation. However, we also prove these sliced metrics are not\textit{not} bi-Lipschitz equivalent to the classical ones in most cases, and also the spaces are (except for an endpoint case) not\textit{not} geodesic. The completeness, duality, and non-geodesicness are new even in the sliced and max-sliced Wasserstein cases, and non bi-Lipschitz equivalence is only known for a few specific cases. In particular this indicates that sliced and max-sliced Wasserstein metrics are not suitable direct replacements for the classical Monge--Kantorovich metrics in problems where the specific metric or geodesic structure are critical.

Keywords

Cite

@article{arxiv.2311.15874,
  title  = {Sliced optimal transport: is it a suitable replacement?},
  author = {Jun Kitagawa and Asuka Takatsu},
  journal= {arXiv preprint arXiv:2311.15874},
  year   = {2024}
}

Comments

26 pages, added results on non bi-Lipschitz equivalence when q is infinity, split off results on disintegrated metrics (to be posted separately), changed title. Comments welcome!

R2 v1 2026-06-28T13:32:45.161Z