On adaptive confidence sets for the Wasserstein distances

DOI10.3150/22-BEJ1535arXiv2111.08505OpenAlexW3213879505MaRDI QIDQ6103231FDOQ6103231

Authors: Neil Deo, Thibault Randrianarisoa 

Publication date: 2 June 2023

Published in: Bernoulli (Search for Journal in Brave)

Abstract: In the density estimation model, we investigate the problem of constructing adaptive honest confidence sets with radius measured in Wasserstein distance $W_p$, $pgeq1$, and for densities with unknown regularity measured on a Besov scale. As sampling domains, we focus on the $d-$dimensional torus $mathbb{T}^d$, in which case $1leqpleq2$, and $mathbb{R}^d$, for which $p=1$. We identify necessary and sufficient conditions for the existence of adaptive confidence sets with diameters of the order of the regularity-dependent $W_p$-minimax estimation rate. Interestingly, it appears that the possibility of such adaptation of the diameter depends on the dimension of the underlying space. In low dimensions, $dleq4$, adaptation to any regularity is possible. In higher dimensions, adaptation is possible if and only if the underlying regularities belong to some interval of width at least $d/(d-4)$. This contrasts with the usual $L_p-$theory where, independently of the dimension, adaptation requires regularities to lie in a small fixed-width window. For configurations allowing these adaptive sets to exist, we explicitly construct confidence regions via the method of risk estimation, centred at adaptive estimators. Those are the first results in a statistical approach to adaptive uncertainty quantification with Wasserstein distances. Our analysis and methods extend more globally to weak losses such as Sobolev norm distances with negative smoothness indices.

Full work available at URL: https://arxiv.org/abs/2111.08505

zbMATH Keywords

uncertainty quantificationWasserstein distancenonparametric confidence sets

Mathematics Subject Classification ID

Nonparametric inference (62Gxx) Statistical decision theory (62Cxx) Limit theorems in probability theory (60Fxx)

Cites Work

• Confidence bands in density estimation
• Density estimation by wavelet thresholding
• Mathematical foundations of infinite-dimensional statistical models
• Optimal Transport
• Nonparametric Bernstein-von Mises theorems in Gaussian white noise
• Anti-concentration and honest, adaptive confidence bands
• On the Bernstein-von Mises phenomenon for nonparametric Bayes procedures
• Title not available (Why is that?)
• The Speed of Mean Glivenko-Cantelli Convergence
• Title not available (Why is that?)
• On minimax density estimation on \(\mathbb R\)
• Frequentist coverage of adaptive nonparametric Bayesian credible sets
• On the rate of convergence in Wasserstein distance of the empirical measure
• Central limit theorems for the Wasserstein distance between the empirical and the true distributions
• Adaptive nonparametric confidence sets
• Title not available (Why is that?)
• On nonparametric confidence intervals
• Adaptive confidence interval for pointwise curve estimation.
• Honest adaptive confidence bands and self-similar functions
• Adaptive confidence balls
• A sharp adaptive confidence ball for self-similar functions
• Adaptive confidence sets in \(L^2\)
• Honest and adaptive confidence sets in \(L_p\)
• On adaptive inference and confidence bands
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Asymptotic frequentist coverage properties of Bayesian credible sets for sieve priors
• Adaptive Bernstein-von Mises theorems in Gaussian white noise
• Adaptation bounds for confidence bands under self-similarity
• A review of uncertainty quantification for density estimation

Cited In (2)

• Honest and adaptive confidence sets in \(L_p\)
• Uniform confidence band for optimal transport map on one-dimensional data

This page was built for publication: On adaptive confidence sets for the Wasserstein distances

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6103231)