Maximum-a-Posteriori Estimation with Bayesian Confidence Regions

DOI10.1137/16M1071249zbMATH Open1375.94024arXiv1602.08590OpenAlexW2963891400MaRDI QIDQ5266375FDOQ5266375

Marcelo Pereyra

Publication date: 2 June 2017

Published in: SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Abstract: Solutions to inverse problems that are ill-conditioned or ill-posed may have significant intrinsic uncertainty. Unfortunately, analysing and quantifying this uncertainty is very challenging, particularly in high-dimensional problems. As a result, while most modern mathematical imaging methods produce impressive point estimation results, they are generally unable to quantify the uncertainty in the solutions delivered. This paper presents a new general methodology for approximating Bayesian high-posterior-density credibility regions in inverse problems that are convex and potentially very high-dimensional. The approximations are derived by using recent concentration of measure results related to information theory for log-concave random vectors. A remarkable property of the approximations is that they can be computed very efficiently, even in large-scale problems, by using standard convex optimisation techniques. In particular, they are available as a by-product in problems solved by maximum-a-posteriori estimation. The approximations also have favourable theoretical properties, namely they outer-bound the true high-posterior-density credibility regions, and they are stable with respect to model dimension. The proposed methodology is illustrated on two high-dimensional imaging inverse problems related to tomographic reconstruction and sparse deconvolution, where the approximations are used to perform Bayesian hypothesis tests and explore the uncertainty about the solutions, and where proximal Markov chain Monte Carlo algorithms are used as benchmark to compute exact credible regions and measure the approximation error.

Full work available at URL: https://arxiv.org/abs/1602.08590

zbMATH Keywords

Bayesian inferenceconvex optimizationuncertainty quantificationinverse problemsstatistical image processing

Mathematics Subject Classification ID

Bayesian inference (62F15) Bayesian problems; characterization of Bayes procedures (62C10) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Computing methodologies for image processing (68U10) Inverse problems in optimal control (49N45)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Phase retrieval via matrix completion
• Statistical and computational inverse problems.
• Simultaneous statistical inference. 2nd ed
• Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
• An Augmented Lagrangian Approach to the Constrained Optimization Formulation of Imaging Inverse Problems
• Compressed sensing
• Proximal Splitting Methods in Signal Processing
• Dictionary learning for noisy and incomplete hyperspectral images
• Alternating Direction Method for Image Inpainting in Wavelet Domains
• The Bernstein-von Mises theorem and nonregular models
• Empirical Bayes Confidence Intervals Based on Bootstrap Samples
• Learning Multiscale Sparse Representations for Image and Video Restoration
• Compressive Blind Image Deconvolution
• Variational Bayesian Super Resolution
• Concentration of the information in data with log-concave distributions
• The Entropy Per Coordinate of a Random Vector is Highly Constrained Under Convexity Conditions
• Super-resolution of positive sources: the discrete setup
• The Variational Gaussian Approximation Revisited
• A nonlocal Bayesian image denoising algorithm
• Large‐Scale Inverse Problems and Quantification of Uncertainty
• Sparse and Redundant Modeling of Image Content Using an Image-Signature-Dictionary
• A new semiblind deconvolution approach for Fourier-based image restoration: an application in astronomy
• Bayesian computation: a summary of the current state, and samples backwards and forwards
• Proximal Markov chain Monte Carlo algorithms
• Photographing Paintings by Image Fusion

Cited In (15)

• Efficient Bayesian Computation for Low-Photon Imaging Problems
• Bayesian imaging inverse problem with SA-roundtrip prior via HMC-pCN sampler
• Efficient Bayesian Computation by Proximal Markov Chain Monte Carlo: When Langevin Meets Moreau
• Patch-Based Image Restoration Using Expectation Propagation
• Normalizing constants of log-concave densities
• On maximum a posteriori estimation with Plug \& Play priors and stochastic gradient descent
• Accelerated Bayesian imaging by relaxed proximal-point Langevin sampling
• Equivalence of weak and strong modes of measures on topological vector spaces
• Sparse Online Variational Bayesian Regression
• The Split Gibbs Sampler Revisited: Improvements to Its Algorithmic Structure and Augmented Target Distribution
• Scalable Bayesian Uncertainty Quantification in Imaging Inverse Problems via Convex Optimization
• Bayesian mode and maximum estimation and accelerated rates of contraction
• A Proximal Markov Chain Monte Carlo Method for Bayesian Inference in Imaging Inverse Problems: When Langevin Meets Moreau
• Bayesian Inference and Uncertainty Quantification for Medical Image Reconstruction with Poisson Data
• Accelerating Proximal Markov Chain Monte Carlo by Using an Explicit Stabilized Method

Uses Software

• UNLocBoX
• TwIST
• PhaseLift

This page was built for publication: Maximum-a-Posteriori Estimation with Bayesian Confidence Regions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5266375)