The Observable Wasserstein Distance

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Santos, Edivaldo Lopes dos, Mauri, Leandro Vicente, Mio, Washington, Needham, Tom
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917479651999744
author Santos, Edivaldo Lopes dos
Mauri, Leandro Vicente
Mio, Washington
Needham, Tom
author_facet Santos, Edivaldo Lopes dos
Mauri, Leandro Vicente
Mio, Washington
Needham, Tom
contents We introduce the observable Wasserstein distance, a framework for deriving lower bounds on the Wasserstein distance between probability measures on Polish metric spaces, designed to bypass the computational intractability of exact optimal transport in large-scale, non-Euclidean datasets. Analogous to the sliced Wasserstein distance in $\mathbb{R}^d$, our approach projects measures onto the real line via 1-Lipschitz observables and computes the Wasserstein distances between the resulting pushforward distributions. We define a hierarchy of pseudo-metrics by restricting observables to a nested chain of subspaces. A central theoretical contribution is an injectivity result linking the metric covering dimension of the support of a measure to the specific order in the hierarchy that guarantees unique recovery. This serves as a metric-space analogue to the Cramér-Wold Device for Euclidean distributions. We demonstrate that this hierarchy offers a tunable trade-off between sharpness as a lower bound on the Wasserstein distance and computational efficiency. We also present a discrete computational model for finite grids and numerical experiments validating the efficacy and utility of these approximations.
format Preprint
id arxiv_https___arxiv_org_abs_2605_09916
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Observable Wasserstein Distance
Santos, Edivaldo Lopes dos
Mauri, Leandro Vicente
Mio, Washington
Needham, Tom
Metric Geometry
Machine Learning
We introduce the observable Wasserstein distance, a framework for deriving lower bounds on the Wasserstein distance between probability measures on Polish metric spaces, designed to bypass the computational intractability of exact optimal transport in large-scale, non-Euclidean datasets. Analogous to the sliced Wasserstein distance in $\mathbb{R}^d$, our approach projects measures onto the real line via 1-Lipschitz observables and computes the Wasserstein distances between the resulting pushforward distributions. We define a hierarchy of pseudo-metrics by restricting observables to a nested chain of subspaces. A central theoretical contribution is an injectivity result linking the metric covering dimension of the support of a measure to the specific order in the hierarchy that guarantees unique recovery. This serves as a metric-space analogue to the Cramér-Wold Device for Euclidean distributions. We demonstrate that this hierarchy offers a tunable trade-off between sharpness as a lower bound on the Wasserstein distance and computational efficiency. We also present a discrete computational model for finite grids and numerical experiments validating the efficacy and utility of these approximations.
title The Observable Wasserstein Distance
topic Metric Geometry
Machine Learning
url https://arxiv.org/abs/2605.09916