Lipschitz continuity of probability kernels in the optimal transport framework

DOI10.1214/23-AIHP1389zbMATH Open1530.49041arXiv2010.08380OpenAlexW3092864612MaRDI QIDQ6187900FDOQ6187900

Authors: Emanuele Dolera, Edoardo Mainini 

Publication date: 16 January 2024

Published in: Annales de l'Institut Henri Poincaré. Probabilités et Statistiques (Search for Journal in Brave)

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.

Full work available at URL: https://arxiv.org/abs/2010.08380

Recommendations

• On the local Lipschitz stability of Bayesian inverse problems
• Generalized Lipschitz-continuity of integrals with respect to a parameter of the intergrating probability measure
• Quantitative stability of optimal transport maps under variations of the target measure
• On the well-posedness of Bayesian inverse problems
• Continuity properties of the extension of a locally Lipschitz continuous map to the space of probability measures

zbMATH Keywords

optimal transportWasserstein distanceBayesian consistencyBenamou-Brenier formulaprobability kernelBayes formula

Mathematics Subject Classification ID

Bayesian inference (62F15) Optimal transportation (49Q22) Large deviations (60F10)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• On Choosing and Bounding Probability Metrics
• Gradient flows in metric spaces and in the space of probability measures
• Fundamentals of nonparametric Bayesian inference
• On the consistency of Bayes estimates
• Asymptotic approximations for probability integrals
• Title not available (Why is that?)
• A computational fluid mechanics solution to the Monge-Kantorovich mass transfer problem
• Measure theory and fine properties of functions
• Robust Bayesian inference via coarsening
• Variation and optimization of formes. A geometric analysis
• A dozen de Finetti-style results in search of a theory
• Conditional distributions as derivatives
• Best constant in Sobolev inequality
• A note on the Poincaré inequality for convex domains
• An introduction to continuum mechanics
• THE GEOMETRY OF DISSIPATIVE EVOLUTION EQUATIONS: THE POROUS MEDIUM EQUATION
• Inverse problems: a Bayesian perspective
• The local regularity of solutions of degenerate elliptic equations
• An optimal Poincaré inequality in $L^1$ for convex domains
• Title not available (Why is that?)
• Title not available (Why is that?)
• A first course in Sobolev spaces
• An introduction to Bayesian analysis. Theory and methods.
• An optimal Poincaré inequality for convex domains
• Problèmes isoperimetriques et espaces de Sobolev
• Isoperimetric and analytic inequalities for log-concave probability measures
• Sobolev-Poincaré implies John
• Rate of convergence for ergodic continuous Markov processes: Lyapunov versus Poincaré
• On the rate of convergence in Wasserstein distance of the empirical measure
• The Wiener test for degenerate elliptic equations
• Generalization of an inequality by Talagrand and links with the logarithmic Sobolev inequality
• Logarithmic Sobolev inequalities and stochastic Ising models
• Entropies, convexity, and functional inequalities: on \(\Phi\)-entropies and \(\Phi\)-Sobolev inequalities
• On logarithmic Sobolev inequalities. With a preface of Dominique Bakry and Michel Ledoux
• Conditioning as disintegration
• Best constants in Poincar\'e inequalities for convex domains
• A simple proof of the Poincaré inequality for a large class of probability measures
• Nonsmooth calculus
• Title not available (Why is that?)
• Caratterizzazioni delle tracce sulla frontiera relative ad alcune classi di funzioni in \(n\) variabili
• A remark on optimal weighted Poincaré inequalities for convex domains
• On a new axiomatic theory of probability
• Title not available (Why is that?)
• Gaussian null sets and differentiability of Lipschitz map on Banach spaces
• Approximation of laws of random probabilities by mixtures of Dirichlet distributions with applications to nonparametric Bayesian inference
• Frequentistic approximations to Bayesian prevision of exchangeable random elements
• One-dimensional empirical measures, order statistics, and Kantorovich transport distances
• On Fréchet differentiability of Lipschitz maps between Banach spaces
• Note on Friedrichs' inequality in \(N\)-star-shaped domains
• On Rates of Convergence for Bayesian Density Estimation
• Poincaré and logarithmic Sobolev inequalities by decomposition of the energy landscape
• Bayesian inverse problems for functions and applications to fluid mechanics
• Continuous versions of regular conditional distributions
• Approximating priors by finite mixtures of conjugate distributions for an exponential family
• Well-posed Bayesian geometric inverse problems arising in subsurface flow
• Title not available (Why is that?)
• Continuous disintegrations of Gaussian processes
• Hypercontractivity for perturbed diffusion semigroups
• On the well-posedness of Bayesian inverse problems
• Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors
• Convex Sobolev inequalities derived from entropy dissipation
• The Bayesian formulation and well-posedness of fractional elliptic inverse problems
• A Sobolev inequality and Neumann heat kernel estimate for unbounded domains
• Unbounded Poincaré domains
• Empirical Bayes deconvolution estimates
• On computability and disintegration
• Uniform rates of the Glivenko-Cantelli convergence and their use in approximating Bayesian inferences
• On uniform continuity of posterior distributions
• Note on ``Frequentistic approximations to Bayesian prevision of exchangeable random elements

This page was built for publication: Lipschitz continuity of probability kernels in the optimal transport framework

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6187900)