Dimension-independent likelihood-informed MCMC

Abstract

Many Bayesian inference problems require exploring the posterior distribution of high-dimensional parameters that represent the discretization of an underlying function. This work introduces a family of Markov chain Monte Carlo (MCMC) samplers that can adapt to the particular structure of a posterior distribution over functions. Two distinct lines of research intersect in the methods developed here. First, we introduce a general class of operator-weighted proposal distributions that are well defined on function space, such that the performance of the resulting MCMC samplers is independent of the discretization of the function. Second, by exploiting local Hessian information and any associated low-dimensional structure in the change from prior to posterior distributions, we develop an inhomogeneous discretization scheme for the Langevin stochastic differential equation that yields operator-weighted proposals adapted to the non-Gaussian structure of the posterior. The resulting dimension-independent, likelihood-informed (DILI) MCMC samplers may be useful for a large class of high-dimensional problems where the target probability measure has a density with respect to a Gaussian reference measure. Two nonlinear inverse problems are used to demonstrate the efficiency of these DILI samplers: an elliptic PDE coefficient inverse problem and path reconstruction in a conditioned diffusion.

References in corpus (1)

• Likelihood-informed dimension reduction for nonlinear inverse problems

Cited by in corpus (59)

• Likelihood-informed dimension reduction for nonlinear inverse problems
• An introduction to sampling via measure transport
• Geometric MCMC for Infinite-Dimensional Inverse Problems
• A practical and efficient approach for Bayesian quantum state estimation
• Bayesian Parameter Estimation for Dynamical Models in Systems Biology
• Improving Simulation Efficiency of MCMC for Inverse Modeling of Hydrologic Systems with a Kalman-Inspired Proposal Distribution
• 2022 Review of Data-Driven Plasma Science
• On a generalization of the preconditioned Crank-Nicolson Metropolis algorithm
• Accelerating MCMC with active subspaces
• Scalable posterior approximations for large-scale Bayesian inverse problems via likelihood-informed parameter and state reduction
• Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems
• A Stein variational Newton method
• Bayesian calibration and sensitivity analysis for a karst aquifer model using active subspaces
• A TV-Gaussian prior for infinite-dimensional Bayesian inverse problems and its numerical implementations
• Residual-based error correction for neural operator accelerated infinite-dimensional Bayesian inverse problems
• Survey of multifidelity methods in uncertainty propagation, inference, and optimization
• Bayesian inference of heterogeneous epidemic models: Application to COVID-19 spread accounting for long-term care facilities
• Image Inversion and Uncertainty Quantification for Constitutive Laws of Pattern Formation
• Bayesian Poroelastic Aquifer Characterization from InSAR Surface Deformation Data Part II: Quantifying the Uncertainty
• Randomized Truncated SVD Levenberg-Marquardt Approach to Geothermal Natural State and History Matching
• Dimension-Robust MCMC in Bayesian Inverse Problems
• Adaptive Dimension Reduction to Accelerate Infinite-Dimensional Geometric Markov Chain Monte Carlo
• Data-Free Likelihood-Informed Dimension Reduction of Bayesian Inverse Problems
• Accelerated dimension-independent adaptive Metropolis
• Variational Bayes' method for functions with applications to some inverse problems
• Efficient parameter estimation for a methane hydrate model with active subspaces
• Generalized Parallel Tempering on Bayesian Inverse Problems
• Prior normalization for certified likelihood-informed subspace detection of Bayesian inverse problems
• Fast Gibbs sampling for high-dimensional Bayesian inversion
• hIPPYlib: An Extensible Software Framework for Large-Scale Inverse Problems Governed by PDEs; Part I: Deterministic Inversion and Linearized Bayesian Inference
• Randomized maximum likelihood based posterior sampling
• Data-Driven Forward Discretizations for Bayesian Inversion
• Stein variational gradient descent on infinite-dimensional space and applications to statistical inverse problems
• Log-Gaussian Gamma Processes for Training Bayesian Neural Networks in Raman and CARS Spectroscopies
• Multilevel Dimension-Independent Likelihood-Informed MCMC for Large-Scale Inverse Problems
• Ensemble sampler for infinite-dimensional inverse problems
• Deep Markov Chain Monte Carlo
• Iterative Construction of Gaussian Process Surrogate Models for Bayesian Inference
• Adaptive inference over Besov spaces in the white noise model using $p$-exponential priors
• An adaptive independence sampler MCMC algorithm for infinite dimensional Bayesian inferences
• Finite Element Representations of Gaussian Processes: Balancing Numerical and Statistical Accuracy
• A Bayesian Approach for Inferring Sea Ice Loads
• Scalable optimization-based sampling on function space
• Scaling Up Bayesian Uncertainty Quantification for Inverse Problems using Deep Neural Networks
• On an adaptive preconditioned Crank-Nicolson MCMC algorithm for infinite dimensional Bayesian inferences
• Analysis of sloppiness in model simulations: unveiling parameter uncertainty when mathematical models are fitted to data
• Iterative importance sampling algorithms for parameter estimation
• Goal-oriented optimal approximations of Bayesian linear inverse problems
• Bayesian inverse problems with $l_1$ priors: a Randomize-then-Optimize approach
• On Unifying Randomized Methods For Inverse Problems
• MALA-within-Gibbs samplers for high-dimensional distributions with sparse conditional structure
• Spatial localization for nonlinear dynamical stochastic models for excitable media
• Bayesian identification of discontinuous fields with an ensemble-based variable separation multiscale method
• A posteriori stochastic correction of reduced models in delayed acceptance MCMC, with application to multiphase subsurface inverse problems
• Localization for MCMC: sampling high-dimensional posterior distributions with local structure
• hIPPYlib-MUQ: A Bayesian Inference Software Framework for Integration of Data with Complex Predictive Models under Uncertainty
• Optimal low-rank posterior mean and distribution approximation in linear Gaussian inverse problems on Hilbert spaces
• Optimization-Based MCMC Methods for Nonlinear Hierarchical Statistical Inverse Problems
• Sequential Ensemble Transform for Bayesian Inverse Problems