Maximum-A-Posteriori Estimates in Linear Inverse Problems with Log-concave Priors are Proper Bayes Estimators

Abstract

A frequent matter of debate in Bayesian inversion is the question, which of the two principle point-estimators, the maximum-a-posteriori (MAP) or the conditional mean (CM) estimate is to be preferred. As the MAP estimate corresponds to the solution given by variational regularization techniques, this is also a constant matter of debate between the two research areas. Following a theoretical argument - the Bayes cost formalism - the CM estimate is classically preferred for being the Bayes estimator for the mean squared error cost while the MAP estimate is classically discredited for being only asymptotically the Bayes estimator for the uniform cost function. In this article we present recent theoretical and computational observations that challenge this point of view, in particular for high-dimensional sparsity-promoting Bayesian inversion. Using Bregman distances, we present new, proper convex Bayes cost functions for which the MAP estimator is the Bayes estimator. We complement this finding by results that correct further common misconceptions about MAP estimates. In total, we aim to rehabilitate MAP estimates in linear inverse problems with log-concave priors as proper Bayes estimators.

References in corpus (2)

• Discretization-invariant Bayesian inversion and Besov space priors
• Hierarchical models in statistical inverse problems and the Mumford--Shah functional

Cited by in corpus (12)

• Deep Bayesian Inversion
• D${}^3$PO - Denoising, Deconvolving, and Decomposing Photon Observations
• Generalized modes in Bayesian inverse problems
• Solution Paths of Variational Regularization Methods for Inverse Problems
• Variational Bayes' method for functions with applications to some inverse problems
• Denoising, deconvolving and decomposing multi-domain photon observations- The D4PO algorithm
• Stein variational gradient descent on infinite-dimensional space and applications to statistical inverse problems
• Maximum a posteriori probability estimates in infinite-dimensional Bayesian inverse problems
• Bregman Cost for Non-Gaussian Noise
• Physics-Informed Machine Learning with Conditional Karhunen-Loève Expansions
• Diffusion tensor imaging with deterministic error bounds
• Kinetic Methods for Inverse Problems