Multilevel sequential Monte Carlo for Bayesian inverse problems

DOI10.1016/J.JCP.2018.04.014zbMATH Open1392.65012arXiv1709.09763OpenAlexW2759608098WikidataQ129910929 ScholiaQ129910929MaRDI QIDQ725440FDOQ725440

Authors: Jonas Latz, Iason Papaioannou, Elisabeth Ullmann 

Publication date: 1 August 2018

Published in: Journal of Computational Physics (Search for Journal in Brave)

Abstract: The identification of parameters in mathematical models using noisy observations is a common task in uncertainty quantification. We employ the framework of Bayesian inversion: we combine monitoring and observational data with prior information to estimate the posterior distribution of a parameter. Specifically, we are interested in the distribution of a diffusion coefficient of an elliptic PDE. In this setting, the sample space is high-dimensional, and each sample of the PDE solution is expensive. To address these issues we propose and analyse a novel Sequential Monte Carlo (SMC) sampler for the approximation of the posterior distribution. Classical, single-level SMC constructs a sequence of measures, starting with the prior distribution, and finishing with the posterior distribution. The intermediate measures arise from a tempering of the likelihood, or, equivalently, a rescaling of the noise. The resolution of the PDE discretisation is fixed. In contrast, our estimator employs a hierarchy of PDE discretisations to decrease the computational cost. We construct a sequence of intermediate measures by decreasing the temperature or by increasing the discretisation level at the same time. This idea builds on and generalises the multi-resolution sampler proposed in [P.S. Koutsourelakis, J. Comput. Phys., 228 (2009), pp. 6184-6211] where a bridging scheme is used to transfer samples from coarse to fine discretisation levels. Importantly, our choice between tempering and bridging is fully adaptive. We present numerical experiments in 2D space, comparing our estimator to single-level SMC and the multi-resolution sampler.

Full work available at URL: https://arxiv.org/abs/1709.09763

Recommendations

• Multilevel sequential Monte Carlo samplers
• Multilevel sequential Monte Carlo with dimension-independent likelihood-informed proposals
• Sequential Monte Carlo methods for Bayesian elliptic inverse problems
• A Bayesian approach to multiscale inverse problems using the sequential Monte Carlo method
• Multilevel Sequential Monte Carlo Samplers for Normalizing Constants

zbMATH Keywords

particle filteruncertainty quantificationfinite element methodsequential importance samplingpartial differential equationtempering

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Bayesian problems; characterization of Bayes procedures (62C10) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) PDEs in connection with geophysics (35Q86)

Cites Work

• Sequential Monte Carlo Samplers
• Title not available (Why is that?)
• Spatial variation. 2nd ed
• Statistical and computational inverse problems.
• Title not available (Why is that?)
• Monte Carlo sampling methods using Markov chains and their applications
• Fundamentals of Nonparametric Bayesian Inference
• Can local particle filters beat the curse of dimensionality?
• A sequential particle filter method for static models
• Equation of State Calculations by Fast Computing Machines
• Simulating normalizing constants: From importance sampling to bridge sampling to path sampling
• Importance sampling: intrinsic dimension and computational cost
• Inverse problems: a Bayesian perspective
• Multilevel Monte Carlo Path Simulation
• Title not available (Why is that?)
• Probability theory. A comprehensive course
• Inference for Lévy-driven stochastic volatility models via adaptive sequential Monte Carlo
• On the convergence of adaptive sequential Monte Carlo methods
• Sequential Monte Carlo Methods for High-Dimensional Inverse Problems: A Case Study for the Navier--Stokes Equations
• MCMC methods for functions: modifying old algorithms to make them faster
• Dimensionality reduction and polynomial chaos acceleration of Bayesian inference in inverse problems
• Ensemble Kalman methods for inverse problems
• Solving large-scale PDE-constrained Bayesian inverse problems with Riemann manifold Hamiltonian Monte Carlo
• MCMC METHODS FOR DIFFUSION BRIDGES
• Sequential Monte Carlo methods for Bayesian elliptic inverse problems
• A multi-resolution, non-parametric, Bayesian framework for identification of spatially-varying model parameters
• Uncertainty Quantification and Weak Approximation of an Elliptic Inverse Problem
• Multilevel Monte Carlo Methods
• Multilevel sequential Monte Carlo: Mean square error bounds under verifiable conditions
• A Hierarchical Multilevel Markov Chain Monte Carlo Algorithm with Applications to Uncertainty Quantification in Subsurface Flow
• Multilevel sequential Monte Carlo samplers
• Multilevel ensemble Kalman filtering
• An Inverse Problem for the Steady State Diffusion Equation
• Multilevel Estimation of Rare Events
• Iterative updating of model error for Bayesian inversion
• Analysis of the Ensemble Kalman Filter for Inverse Problems
• Multilevel Sequential Monte Carlo Samplers for Normalizing Constants

Cited In (28)

• Bayesian updating and marginal likelihood estimation by cross entropy based importance sampling
• Multifidelity multilevel Monte Carlo to accelerate approximate Bayesian parameter inference for partially observed stochastic processes
• A transport-based multifidelity preconditioner for Markov chain Monte Carlo
• Multilevel Sequential Importance Sampling for Rare Event Estimation
• Bayesian inversion by parallel interacting Markov chains
• A Practical Example for the Non-linear Bayesian Filtering of Model Parameters
• The Ensemble Kalman Filter for Rare Event Estimation
• Sequential ensemble transform for Bayesian inverse problems
• Bayesian Parameter Identification in Cahn--Hilliard Models for Biological Growth
• Transform-based particle filtering for elliptic Bayesian inverse problems
• Multi-index ensemble Kalman filtering
• Multilevel Adaptive Sparse Leja Approximations for Bayesian Inverse Problems
• Bayesian Inverse Problems Are Usually Well-Posed
• Multilevel Markov Chain Monte Carlo for Bayesian Inversion of Parabolic Partial Differential Equations under Gaussian Prior
• Multi-agent reinforcement learning aided sampling algorithms for a class of multiscale inverse problems
• Binned Multilevel Monte Carlo for Bayesian Inverse Problems with Large Data
• Adaptive regularisation for ensemble Kalman inversion
• Analysis of a Class of Multilevel Markov Chain Monte Carlo Algorithms Based on Independent Metropolis–Hastings
• Multilevel higher-order quasi-Monte Carlo Bayesian estimation
• Multilevel Markov Chain Monte Carlo
• Generalized parallel tempering on Bayesian inverse problems
• A multi-resolution, non-parametric, Bayesian framework for identification of spatially-varying model parameters
• Certified Dimension Reduction for Bayesian Updating with the Cross-Entropy Method
• Context-Aware Surrogate Modeling for Balancing Approximation and Sampling Costs in Multifidelity Importance Sampling and Bayesian Inverse Problems
• Advanced Multilevel Monte Carlo Methods
• Bayesian inference of random fields represented with the Karhunen-Loève expansion
• Multilevel Monte Carlo Covariance Estimation for the Computation of Sobol' Indices
• Central limit theorems for multilevel Monte Carlo methods

This page was built for publication: Multilevel sequential Monte Carlo for Bayesian inverse problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q725440)