Lipschitz continuity of probability kernels in the optimal transport framework

TL;DR

This paper establishes conditions under which probability kernels in Bayesian statistics are Lipschitz continuous within the optimal transport framework, ensuring stability of posterior distributions with respect to data errors.

Contribution

It provides new general conditions and explicit bounds for Lipschitz continuity of probability kernels, including in infinite-dimensional and non-dominated cases, within the optimal transport setting.

Findings

Lipschitz continuity results with explicit bounds for finite-dimensional dominated kernels

Extensions to kernels with moving support and infinite-dimensional spaces

Applications to posterior approximation and consistency in Bayesian statistics

Abstract

In Bayesian statistics, a continuity property of the posterior distribution with respect to the observable variable is crucial as it expresses well-posedness, i.e., stability with respect to errors in the measurement of data. Essentially, this requires to analyze the continuity of a probability kernel or, equivalently, of a conditional probability distribution with respect to the conditioning variable. Here, we give general conditions for the Lipschitz continuity of probability kernels with respect to metric structures arising within the optimal transport framework, such as the Wasserstein metric. For dominated probability kernels over finite-dimensional spaces, we show Lipschitz continuity results with a Lipschitz constant enjoying explicit bounds in terms of Fisher-information functionals and weighted Poincar\'e constants. We also provide results for kernels with moving support, for infinite-dimensional spaces and for non dominated kernels. We show applications to several problems in Bayesian statistics, such as approximation of posterior distributions by mixtures and posterior consistency.