Adaptive particle-based approximations of the Gibbs posterior for inverse problems
Preprint
Abstract
In this work, we adopt a general framework based on the Gibbs posterior to update belief distributions for inverse problems governed by partial differential equations (PDEs). The Gibbs posterior formulation is a generalization of standard Bayesian inference that only relies on a loss function connecting the unknown parameters to the data. It is particularly useful when the true data generating mechanism (or noise distribution) is unknown or difficult to specify. The Gibbs posterior coincides with Bayesian updating when a true likelihood function is known and the loss function corresponds to the negative log-likelihood, yet provides subjective inference in more general settings. We employ a sequential Monte Carlo (SMC) approach to approximate the Gibbs posterior using particles. To manage the computational cost of propagating increasing numbers of particles through the loss function, we employ a recently developed local reduced basis method to build an efficient surrogate loss function that is used in the Gibbs update formula in place of the true loss. We derive error bounds for our approximation and propose an adaptive approach to construct the surrogate model in an efficient manner. We demonstrate the efficiency of our approach through several numerical examples.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- An adaptive local reduced basis method for solving PDEs with uncertain inputs and evaluating riskComputer Methods in Applied Mechanics and Engineering, 2019
- Sequential Monte Carlo methods for Bayesian elliptic inverse problemsStatistics and Computing, 2015
- Sequential Monte Carlo Methods for High-Dimensional Inverse Problems: A Case Study for the Navier--Stokes EquationsSIAM/ASA Journal on Uncertainty Quantification, 2014
- A Computational Framework for Infinite-Dimensional Bayesian Inverse Problems Part I: The Linearized Case, with Application to Global Seismic InversionSIAM Journal on Scientific Computing, 2013
- A method for solving stochastic equations by reduced order models and local approximationsJournal of Computational Physics, 2012
- A novel Bayesian strategy for the identification of spatially varying material properties and model validation: an application to static elastographyInternational Journal for Numerical Methods in Engineering, 2012
- Sequential Monte Carlo SamplersJournal of the Royal Statistical Society Series B: Statistical Methodology, 2006
- A Bayesian inference approach to the inverse heat conduction problemInternational Journal of Heat and Mass Transfer, 2004
- A simple method using Morozov's discrepancy principle for solving inverse scattering problemsInverse Problems, 1997
- The use of Morozov's discrepancy principle for Tikhonov regularization for solving nonlinear ill-posed problemsComputing, 1993