MAP estimators for piecewise continuous inversion

DOI10.1088/0266-5611/32/10/105003MaRDI QIDQ2832559zbMATH OpenOpenAlexFDO

Authors M. M. Dunlop, A. M. Stuart

Publication date 11 November 2016

Published in Inverse Problems (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/1509.03136

zbMATH Keywords

Bayesian inversionelectrical impedance tomographygeometric priorsgroundwater flowinverse problemsMAP estimator

Mathematics Subject Classification ID

Numerical solution to inverse problems in abstract spaces (65J22)

Abstract: We study the inverse problem of estimating a field $u$ from data comprising a finite set of nonlinear functionals of $u$, subject to additive noise; we denote this observed data by $y$. Our interest is in the reconstruction of piecewise continuous fields in which the discontinuity set is described by a finite number of geometric parameters. Natural applications include groundwater flow and electrical impedance tomography. We take a Bayesian approach, placing a prior distribution on $u$ and determining the conditional distribution on $u$ given the data $y$. It is then natural to study maximum a posterior (MAP) estimators. Recently (Dashti et al 2013) it has been shown that MAP estimators can be characterised as minimisers of a generalised Onsager-Machlup functional, in the case where the prior measure is a Gaussian random field. We extend this theory to a more general class of prior distributions which allows for piecewise continuous fields. Specifically, the prior field is assumed to be piecewise Gaussian with random interfaces between the different Gaussians defined by a finite number of parameters. We also make connections with recent work on MAP estimators for linear problems and possibly non-Gaussian priors (Helin, Burger 2015) which employs the notion of Fomin derivative. In showing applicability of our theory we focus on the groundwater flow and EIT models, though the theory holds more generally. Numerical experiments are implemented for the groundwater flow model, demonstrating the feasibility of determining MAP estimators for these piecewise continuous models, but also that the geometric formulation can lead to multiple nearby (local) MAP estimators. We relate these MAP estimators to the behaviour of output from MCMC samples of the posterior, obtained using a state-of-the-art function space Metropolis-Hastings method.

Review text

This work discusses the inverse problem of estimating a field from a finite set of nonlinear functionals of the state variable. The focus is on reconstructing piecewise continuous state variable, where the discontinuity is described by a finite number of geometric parameters. The setting covers groundwater flow and electrical impedance tomography. The authors employ a Bayesian formulation with a piecewise continuous fields, and analyze the MAP estimators, which are minimizers of a generalized Onsager-Machlup functional. The theory is illustrated with the groundwater flow, and numerically demonstrated.

Recommendations

• MAP estimators and their consistency in Bayesian nonparametric inverse problems
• Maximum a posteriori probability estimates in infinite-dimensional Bayesian inverse problems
• Maximum a posteriori estimates in linear inverse problems with log-concave priors are proper Bayes estimators
• Well-posed Bayesian geometric inverse problems arising in subsurface flow
• Model distortions in Bayesian MAP reconstruction

Cited in

(17)

• Nonlinear map inversion via state observers
• Infinite-dimensional Bayesian approach for inverse scattering problems of a fractional Helmholtz equation
• Posterior contraction for empirical Bayesian approach to inverse problems under non-diagonal assumption
• Optimization based methods for partially observed chaotic systems
• Sparsity-promoting and edge-preserving maximum a posteriori estimators in non-parametric Bayesian inverse problems
• Equivalence of weak and strong modes of measures on topological vector spaces
• Recursive linearization method for inverse medium scattering problems with complex mixture Gaussian error learning
• Γ -convergence of Onsager–Machlup functionals: I. With applications to maximum a posteriori estimation in Bayesian inverse problems
• Γ-convergence of Onsager–Machlup functionals: II. Infinite product measures on Banach spaces
• Generalized Modes in Bayesian Inverse Problems
• Quasi-Monte Carlo and multilevel Monte Carlo methods for computing posterior expectations in elliptic inverse problems
• MAP estimators and their consistency in Bayesian nonparametric inverse problems
• Variational Bayes' Method for Functions with Applications to Some Inverse Problems
• A Bayesian approach for consistent reconstruction of inclusions
• Simultaneous estimation of piecewise constant coefficients in elliptic pdes via Bayesian level-set methods
• Spatiotemporal Besov Priors for Bayesian Inverse Problems
• The graph limit of the minimizer of the Onsager-Machlup functional and its computation

This page was built for publication: MAP estimators for piecewise continuous inversion

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2832559)