An Improved Central Limit Theorem and Fast Convergence Rates for Entropic Transportation Costs

DOI10.1137/22M149260XarXiv2204.09105OpenAlexW4384927756WikidataQ122122950 ScholiaQ122122950MaRDI QIDQ6136231FDOQ6136231

Authors: Eustasio Del Barrio, Alberto González Sanz, Jean-Michel Loubes, Jonathan Niles-Weed 

Publication date: 29 August 2023

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Abstract: We prove a central limit theorem for the entropic transportation cost between subgaussian probability measures, centered at the population cost. This is the first result which allows for asymptotically valid inference for entropic optimal transport between measures which are not necessarily discrete. In the compactly supported case, we complement these results with new, faster, convergence rates for the expected entropic transportation cost between empirical measures. Our proof is based on strengthening convergence results for dual solutions to the entropic optimal transport problem.

Full work available at URL: https://arxiv.org/abs/2204.09105

zbMATH Keywords

central limit theoremoptimal transportentropic regularizationSinkhorn divergence

Mathematics Subject Classification ID

Multivariate distribution of statistics (62H10) Optimal transportation (49Q22) Central limit and other weak theorems (60F05)

Cites Work

• Title not available (Why is that?)
• Weak convergence and empirical processes. With applications to statistics
• Computational Optimal Transport: With Applications to Data Science
• Mathematical Foundations of Infinite-Dimensional Statistical Models
• Title not available (Why is that?)
• Large Sample Properties of Matching Estimators for Average Treatment Effects
• Wasserstein discriminant analysis
• Distribution-Free Consistent Independence Tests via Center-Outward Ranks and Signs
• I-divergence geometry of probability distributions and minimization problems
• Asymptotics for \(L_2\) functionals of the empirical quantile process, with applications to tests of fit based on weighted Wasserstein distances
• Central limit theorems for the Wasserstein distance between the empirical and the true distributions
• Multivariate Rank-Based Distribution-Free Nonparametric Testing Using Measure Transportation
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Distribution and quantile functions, ranks and signs in dimension \(d\): a measure transportation approach
• Monge-Kantorovich depth, quantiles, ranks and signs
• Bracketing smooth functions
• Robust Wasserstein profile inference and applications to machine learning
• Entropy-regularized 2-Wasserstein distance between Gaussian measures
• Central limit theorems for empirical transportation cost in general dimension
• Multivariate goodness-of-fit tests based on Wasserstein distance
• Entropic optimal transport is maximum-likelihood deconvolution
• Entropic optimal transport: convergence of potentials
• Empirical optimal transport on countable metric spaces: distributional limits and statistical applications
• Inference for Empirical Wasserstein Distances on Finite Spaces
• Minimax estimation of smooth optimal transport maps
• Central limit theorems for entropy-regularized optimal transport on finite spaces and statistical applications
• Uncoupled isotonic regression via minimum Wasserstein deconvolution
• Empirical Regularized Optimal Transport: Statistical Theory and Applications
• Quantitative Stability of Regularized Optimal Transport and Convergence of Sinkhorn's Algorithm
• The Earth Mover's Correlation
• The Sinkhorn algorithm, parabolic optimal transport and geometric Monge-Ampère equations
• On universally consistent and fully distribution-free rank tests of vector independence
• Multivariate ranks and quantiles using optimal transport: consistency, rates and nonparametric testing
• Tackling algorithmic bias in neural-network classifiers using Wasserstein-2 regularization
• Stability of entropic optimal transport and Schrödinger bridges
• A central limit theorem for Lp transportation cost on the real line with application to fairness assessment in machine learning
• Asymptotics for Semidiscrete Entropic Optimal Transport
• Stability of Schrödinger potentials and convergence of Sinkhorn's algorithm
• Two-sample goodness-of-fit tests on the flat torus based on Wasserstein distance and their relevance to structural biology

Cited In (12)

• Diffeomorphic Registration Using Sinkhorn Divergences
• A unifying approach to distributional limits for empirical optimal transport
• Limit distribution theory for smooth \(p\)-Wasserstein distances
• Central limit theorems for semi-discrete Wasserstein distances
• Gromov-Wasserstein distances: entropic regularization, duality and sample complexity
• Monge-Kantorovich superquantiles and expected shortfalls with applications to multivariate risk measurements
• Stability and sample complexity of divergence regularized optimal transport
• Central limit theorems for entropy-regularized optimal transport on finite spaces and statistical applications
• A central limit theorem for Lp transportation cost on the real line with application to fairness assessment in machine learning
• Uniform confidence band for optimal transport map on one-dimensional data
• Statistical inference with regularized optimal transport
• Limit distributions and sensitivity analysis for empirical entropic optimal transport on countable spaces

This page was built for publication: An Improved Central Limit Theorem and Fast Convergence Rates for Entropic Transportation Costs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136231)