Central limit theorems for semi-discrete Wasserstein distances

DOI10.3150/23-BEJ1608MaRDI QIDQ6178576zbMATH OpenFDO

Authors Eustasio Del Barrio, Alberto González Sanz, Jean-Michel Loubes

Publication date 16 January 2024

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2202.06380

zbMATH Keywords

central limit theoremLaguerre cellsoptimal transportoptimal transport potentialssemi-discrete optimal transport

Mathematics Subject Classification ID

Limit theorems in probability theory (60Fxx) Nonparametric inference (62Gxx) Multivariate analysis (62Hxx)

Abstract: We prove a Central Limit Theorem for the empirical optimal transport cost, $sqrtfracnmn+mmathcal{T}_c(P_n,Q_m)-mathcal{T}_c(P,Q)$, in the semi discrete case, i.e when the distribution $P$ is supported in $N$ points, but without assumptions on $Q$. We show that the asymptotic distribution is the supremun of a centered Gaussian process, which is Gaussian under some additional conditions on the probability $Q$ and on the cost. Such results imply the central limit theorem for the $p$-Wassertein distance, for $pgeq1$. This means that, for fixed $N$, the curse of dimensionality is avoided. To better understand the influence of such $N$, we provide bounds of $E|mathcal{W}_1(P,Q_m)-mathcal{W}_1(P,Q)|$ depending on $m$ and $N$. Finally, the semidiscrete framework provides a control on the second derivative of the dual formulation, which yields the first central limit theorem for the optimal transport potentials. The results are supported by simulations that help to visualize the given limits and bounds. We analyse also the cases where classical bootstrap works.

Cites work

• \(L_p\) metrics for compact convex sets
• A central limit theorem for k-means clustering
• A central limit theorem for Lp transportation cost on the real line with application to fairness assessment in machine learning
• A Lagrangian scheme à la Brenier for the incompressible Euler equations
• A note on the regularity of optimal-transport-based center-outward distribution and quantile functions
• An Improved Central Limit Theorem and Fast Convergence Rates for Entropic Transportation Costs
• An optimal Poincaré inequality in L¹ for convex domains
• Boundary regularity of maps with convex potentials. II
• Central limit theorems for empirical transportation cost in general dimension
• Convergence of a Newton algorithm for semi-discrete optimal transport
• Convex Analysis
• Directional differentiability for supremum-type functionals: statistical applications
• Distribution and quantile functions, ranks and signs in dimension \(d\): a measure transportation approach
• Empirical optimal transport on countable metric spaces: distributional limits and statistical applications
• Foundations of quantization for probability distributions
• scientific article; zbMATH DE number 1909499 (Why is no real title available?)
• Hybrid Wasserstein distance and fast distribution clustering
• Inference for empirical Wasserstein distances on finite spaces
• Inference on directionally differentiable functions
• Minimax confidence intervals for the sliced Wasserstein distance
• On how Poincaré inequalities imply weighted ones
• On the role of convexity in isoperimetry, spectral gap and concentration
• Quantitative Stability in the Geometry of Semi-discrete Optimal Transport
• Regularity of monotone transport maps between unbounded domains
• Semi-discrete optimal transport: a solution procedure for the unsquared Euclidean distance case
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• The geometry of optimal transportation
• Two-sample goodness-of-fit tests on the flat torus based on Wasserstein distance and their relevance to structural biology
• Variational Analysis
• Weak convergence and empirical processes. With applications to statistics

Cited in

(10)

• A unifying approach to distributional limits for empirical optimal transport
• Limit distribution theory for smooth \(p\)-Wasserstein distances
• Empirical optimal transport under estimated costs: distributional limits and statistical applications
• Statistical inference with regularized optimal transport
• Convergence of empirical optimal transport in unbounded settings
• Unbalanced Kantorovich-Rubinstein distance, plan, and barycenter on finite spaces: a statistical perspective
• On the uniqueness of Kantorovich potentials
• Stability and statistical inference for semidiscrete optimal transport maps
• On the robustness of semi-discrete optimal transport
• Estimation and inference for the Wasserstein distance between mixing measures in topic models

This page was built for publication: Central limit theorems for semi-discrete Wasserstein distances

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6178576)