Strong equivalence between metrics of Wasserstein type

DOI10.1214/21-ECP383MaRDI QIDQ2064844zbMATH OpenFDO

Authors Erhan Bayraktar, Gaoyue Guo

Publication date 6 January 2022

Published in Electronic Communications in Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1912.08247

zbMATH Keywords

optimal transportWasserstein metricdualitymax-sliced Wasserstein metric

Mathematics Subject Classification ID

Convex programming (90C25)

Abstract: The sliced Wasserstein and more recently max-sliced Wasserstein metrics $m{W}_p$ have attracted abundant attention in data sciences and machine learning due to its advantages to tackle the curse of dimensionality. A question of particular importance is the strong equivalence between these projected Wasserstein metrics and the (classical) Wasserstein metric $W{c}_p$. Recently, Paty and Cuturi have proved the strong equivalence of $m{W}_2$ and $W{c}_2$. We show that the strong equivalence also holds for $p=1$, while we show that the sliced Wasserstein metric does not share this nice property.

Recommendations

• A framework for Wasserstein-1-type metrics
• The equivalence of Fourier-based and Wasserstein metrics on imaging problems
• A class of Wasserstein metrics for probability distributions
• On properties of the generalized Wasserstein distance
• Comparison between \(W_2\) distance and \(\dot{H}^{-1}\) norm, and localization of Wasserstein distance

Cites work

• scientific article; zbMATH DE number 3890132 (Why is no real title available?)
• Approximation by superpositions of a sigmoidal function
• Banach spaces of Lipschitz functions and vector-valued Lipschitz functions
• Computational methods for martingale optimal transport problems
• From Knothe's Rearrangement to Brenier's Optimal Transport Map
• Lectures on analysis on metric spaces
• Mass transportation problems. Vol. 1: Theory. Vol. 2: Applications
• On the optimal map in the 2-dimensional random matching problem
• On the rate of convergence in Wasserstein distance of the empirical measure
• Optimal Transport
• Optimal transport for applied mathematicians. Calculus of variations, PDEs, and modeling
• Some natural subspaces and quotient spaces of \(L^1\)

Cited in

(5)

• A smooth variational principle on Wasserstein space
• A note on the radiant formula and its relations to the sliced Wasserstein distance
• Wasserstein convergence in Bayesian and frequentist deconvolution models
• Statistical inference with regularized optimal transport
• Optimal estimation of high-dimensional Gaussian location mixtures

Describes a project that uses

Uses Software

• Wasserstein GAN

This page was built for publication: Strong equivalence between metrics of Wasserstein type

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2064844)