Gaussian approximation for penalized Wasserstein barycenters

DOI10.3103/S1066530723010039MaRDI QIDQ6044261zbMATH OpenFDO

Authors Nazar Buzun

Publication date 17 May 2023

Published in Mathematical Methods of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1904.00891

zbMATH Keywords

convex analysisGaussian approximationhigh-dimensional statisticsmultivariate central limit theoremWasserstein barycenters

Mathematics Subject Classification ID

Probability theory and stochastic processes (60-XX) Statistics (62-XX)

Abstract: In this work we consider regularized Wasserstein barycenters (average in Wasserstein distance) in Fourier basis. We prove that random Fourier parameters of the barycenter converge to some Gaussian random vector by distribution. The convergence rate has been derived in finite-sample case with explicit dependence on measures count ($n$) and the dimension of parameters ($p$).

Cites work

• A new method for approximations in probability and operator theories
• Barycenters in the Wasserstein space
• Change-point detection in high-dimensional covariance structure
• Concentration inequalities. A nonasymptotic theory of independence
• Convex analysis and monotone operator theory in Hilbert spaces
• Entropic regularization of continuous optimal transport problems
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• Inference for empirical Wasserstein distances on finite spaces
• Large ball probabilities, Gaussian comparison and anti-concentration
• Limit laws of the empirical Wasserstein distance: Gaussian distributions
• On Stein's method for multivariate normal approximation
• On the dependence of the Berry-Esseen bound on dimension
• On the Kantorovich-Rubinstein theorem
• Penalization of barycenters in the Wasserstein space
• Penalized maximum likelihood estimation and effective dimension
• Quadratically regularized optimal transport
• Wasserstein distance, Fourier series and applications

Cited in

(2)

• An inexact PAM method for computing Wasserstein barycenter with unknown supports
• Gradient projection methods for the $n$-coupling problem

This page was built for publication: Gaussian approximation for penalized Wasserstein barycenters

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6044261)