Well-Posed Bayesian Inverse Problems: Priors with Exponential Tails

DOI10.1137/16M1076824zbMATH Open1371.35349arXiv1604.02575WikidataQ115525637 ScholiaQ115525637MaRDI QIDQ5269871FDOQ5269871

Authors: Bamdad Hosseini, Nilima Nigam 

Publication date: 28 June 2017

Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Abstract: We consider the well-posedness of Bayesian inverse problems when the prior measure has exponential tails. In particular, we consider the class of convex (log-concave) probability measures which include the Gaussian and Besov measures as well as certain classes of hierarchical priors. We identify appropriate conditions on the likelihood distribution and the prior measure which guarantee existence, uniqueness and stability of the posterior measure with respect to perturbations of the data. We also consider consistent approximations of the posterior such as discretization by projection. Finally, we present a general recipe for construction of convex priors on Banach spaces which will be of interest in practical applications where one often works with spaces such as $L^2$ or the continuous functions.

Full work available at URL: https://arxiv.org/abs/1604.02575

Recommendations

• On the well-posedness of Bayesian inverse problems
• Well-posed Bayesian inverse problems with infinitely divisible and heavy-tailed prior measures
• Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors
• Bayesian inverse problems with Gaussian priors
• Bayesian inverse problems with non-conjugate priors
• Besov priors for Bayesian inverse problems
• Bayesian Inverse Problems Are Usually Well-Posed
• Inverse problems: a Bayesian perspective
• Inverse problems in the Bayesian framework

zbMATH Keywords

Bayesiannon-Gaussianinverse problemsconvex measure, log-concave distribution

Mathematics Subject Classification ID

Parametric inference (62F99) Probability theory on linear topological spaces (60B11) Inverse problems for PDEs (35R30) Theoretical approximation in context of PDEs (35A35)

Cites Work

• Measure theory. Vol. I and II
• The horseshoe estimator for sparse signals
• Statistical and computational inverse problems.
• Title not available (Why is that?)
• Foundations of Modern Probability
• On the consistency of Bayes estimates
• Convex measures on locally convex spaces
• An introduction to partial differential equations
• Discretization-invariant Bayesian inversion and Besov space priors
• Inverse problems: a Bayesian perspective
• A mathematical introduction to compressive sensing
• Computational Methods for Inverse Problems
• Title not available (Why is that?)
• Introduction to Bayesian Scientific Computing
• On the Asymptotic Behavior of Bayes' Estimates in the Discrete Case
• Partial differential equations. I: Basic theory
• Analysis of the Gibbs Sampler for Hierarchical Inverse Problems
• Title not available (Why is that?)
• Differentiable measures and the Malliavin calculus
• A basis theory primer.
• Log-concave probability and its applications
• Deblurring Images
• MCMC methods for functions: modifying old algorithms to make them faster
• Maximuma posterioriestimates in linear inverse problems with log-concave priors are proper Bayes estimators
• Approximation of Bayesian Inverse Problems for PDEs
• Uncertainty Quantification and Weak Approximation of an Elliptic Inverse Problem
• Besov priors for Bayesian inverse problems
• Well-posed Bayesian geometric inverse problems arising in subsurface flow
• Hierarchical regularization for edge-preserving reconstruction of PET images
• Maximuma posterioriprobability estimates in infinite-dimensional Bayesian inverse problems
• Posterior consistency for Gaussian process approximations of Bayesian posterior distributions
• Sparsity-promoting Bayesian inversion

Cited In (24)

• Inverse stable prior for exponential models
• Constructive deep ReLU neural network approximation
• Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors
• Quasi-Monte Carlo Bayesian estimation under Besov priors in elliptic inverse problems
• Bayesian Inference of an Uncertain Generalized Diffusion Operator
• Accelerated Bayesian imaging by relaxed proximal-point Langevin sampling
• Cauchy Markov random field priors for Bayesian inversion
• Besov priors for Bayesian inverse problems
• Weak and TV consistency in Bayesian uncertainty quantification using disintegration
• Multilevel Monte Carlo FEM for elliptic PDEs with Besov random tree priors
• Discretization-invariant Bayesian inversion and Besov space priors
• Bayesian Inverse Problems Are Usually Well-Posed
• Two Metropolis--Hastings Algorithms for Posterior Measures with Non-Gaussian Priors in Infinite Dimensions
• Bayesian inversion of a diffusion model with application to biology
• A Statistical Approach to Surface Metrology for 3D-Printed Stainless Steel
• A Bayesian scheme for reconstructing obstacles in acoustic waveguides
• Sparse approximation of triangular transports. I: The finite-dimensional case
• On the Well-posedness of Bayesian Inverse Problems
• Spectral gaps and error estimates for infinite-dimensional Metropolis-Hastings with non-Gaussian priors
• Well-Posed Bayesian Inverse Problems with Infinitely Divisible and Heavy-Tailed Prior Measures
• Random Forward Models and Log-Likelihoods in Bayesian Inverse Problems
• On the local Lipschitz stability of Bayesian inverse problems
• Bayesian inference of random fields represented with the Karhunen-Loève expansion
• Error control of the numerical posterior with Bayes factors in Bayesian uncertainty quantification

This page was built for publication: Well-Posed Bayesian Inverse Problems: Priors with Exponential Tails

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5269871)