A computational framework for infinite-dimensional Bayesian inverse problems. II: stochastic Newton MCMC with application to ice sheet flow inverse problems

DOI10.1137/130934805arXiv1308.6221OpenAlexW2962841346MaRDI QIDQ2929997FDOQ2929997

Authors:

Publication date: 17 November 2014

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Abstract: We address the numerical solution of infinite-dimensional inverse problems in the framework of Bayesian inference. In the Part I companion to this paper (arXiv.org:1308.1313), we considered the linearized infinite-dimensional inverse problem. Here in Part II, we relax the linearization assumption and consider the fully nonlinear infinite-dimensional inverse problem using a Markov chain Monte Carlo (MCMC) sampling method. To address the challenges of sampling high-dimensional pdfs arising from Bayesian inverse problems governed by PDEs, we build on the stochastic Newton MCMC method. This method exploits problem structure by taking as a proposal density a local Gaussian approximation of the posterior pdf, whose construction is made tractable by invoking a low-rank approximation of its data misfit component of the Hessian. Here we introduce an approximation of the stochastic Newton proposal in which we compute the low-rank-based Hessian at just the MAP point, and then reuse this Hessian at each MCMC step. We compare the performance of the proposed method to the original stochastic Newton MCMC method and to an independence sampler. The comparison of the three methods is conducted on a synthetic ice sheet inverse problem. For this problem, the stochastic Newton MCMC method with a MAP-based Hessian converges at least as rapidly as the original stochastic Newton MCMC method, but is far cheaper since it avoids recomputing the Hessian at each step. On the other hand, it is more expensive per sample than the independence sampler; however, its convergence is significantly more rapid, and thus overall it is much cheaper. Finally, we present extensive analysis and interpretation of the posterior distribution, and classify directions in parameter space based on the extent to which they are informed by the prior or the observations.

Full work available at URL: https://arxiv.org/abs/1308.6221

Recommendations

• A stochastic Newton MCMC method for large-scale statistical inverse problems with application to seismic inversion
• Dimension-independent likelihood-informed MCMC
• Ensemble sampler for infinite-dimensional inverse problems
• Low-rank independence samplers in hierarchical Bayesian inverse problems
• Solving large-scale PDE-constrained Bayesian inverse problems with Riemann manifold Hamiltonian Monte Carlo

zbMATH Keywords

Bayesian inferenceMCMCuncertainty quantificationlow-rank approximationinfinite-dimensional inverse problemsice sheet dynamicsstochastic Newton

Mathematics Subject Classification ID

Bayesian inference (62F15) Numerical analysis or methods applied to Markov chains (65C40) PDEs in connection with control and optimization (35Q93) Inverse problems for PDEs (35R30) Newton-type methods (49M15) Glaciology (86A40) PDEs in connection with statistics (35Q62)

Cited In (98)

• Taylor approximation and variance reduction for PDE-constrained optimal control under uncertainty
• Pass-efficient randomized algorithms for low-rank matrix approximation using any number of views
• A Bayesian level set method for an inverse medium scattering problem in acoustics
• A Fast and Scalable Computational Framework for Large-Scale High-Dimensional Bayesian Optimal Experimental Design
• Hierarchical off-diagonal low-rank approximation of Hessians in inverse problems, with application to ice sheet model initialization
• Bayesian approach to inverse scattering with topological priors
• A Variational Neural Network Approach for Glacier Modelling with Nonlinear Rheology
• A Randomized Maximum A Posteriori Method for Posterior Sampling of High Dimensional Nonlinear Bayesian Inverse Problems
• Exploration of anisotropic design space by using unified Taylor-cokriging method
• GPU-accelerated particle methods for evaluation of sparse observations for inverse problems constrained by diffusion PDEs
• A Bayesian method for an inverse transmission scattering problem in acoustics
• Solving large-scale PDE-constrained Bayesian inverse problems with Riemann manifold Hamiltonian Monte Carlo
• Multifidelity deep operator networks for data-driven and physics-informed problems
• Low-rank independence samplers in hierarchical Bayesian inverse problems
• Stein variational reduced basis Bayesian inversion
• Taylor approximation for chance constrained optimization problems governed by partial differential equations with high-dimensional random parameters
• PDE-constrained optimization in medical image analysis
• Iterative importance sampling algorithms for parameter estimation
• Accurate solution of Bayesian inverse uncertainty quantification problems combining reduced basis methods and reduction error models
• A meshfree approach to non-Newtonian free surface ice flow: application to the Haut Glacier d'Arolla
• Variational inference at glacier scale
• The statistical finite element method (statFEM) for coherent synthesis of observation data and model predictions
• Weak and TV consistency in Bayesian uncertainty quantification using disintegration
• Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems
• Dimension-independent likelihood-informed MCMC
• Scalable and efficient algorithms for the propagation of uncertainty from data through inference to prediction for large-scale problems, with application to flow of the antarctic ice sheet
• Dynamically coupling the non-linear Stokes equations with the shallow ice approximation in glaciology: description and first applications of the ISCAL method
• Data-free likelihood-informed dimension reduction of Bayesian inverse problems
• Learning physics-based models from data: perspectives from inverse problems and model reduction
• Generalized Modes in Bayesian Inverse Problems
• On the implementation of a quasi-Newton interior-point method for PDE-constrained optimization using finite element discretizations
• Estimation of the Robin coefficient field in a Poisson problem with uncertain conductivity field
• A stochastic subspace approach to gradient-free optimization in high dimensions
• Spatial localization for nonlinear dynamical stochastic models for excitable media
• Localization for MCMC: sampling high-dimensional posterior distributions with local structure
• Goal-oriented optimal approximations of Bayesian linear inverse problems
• Parallel local approximation MCMC for expensive models
• A Bayesian level set method for the shape reconstruction of inverse scattering problems in elasticity
• MALA-within-Gibbs samplers for high-dimensional distributions with sparse conditional structure
• Mixing rates for Hamiltonian Monte Carlo algorithms in finite and infinite dimensions
• A Bayesian Approach to Estimating Background Flows from a Passive Scalar
• A path-integral approach to Bayesian inference for inverse problems using the semiclassical approximation
• Accelerating the Bayesian inference of inverse problems by using data-driven compressive sensing method based on proper orthogonal decomposition
• Statistical treatment of inverse problems constrained by differential equations-based models with stochastic terms
• Tensor train construction from tensor actions, with application to compression of large high order derivative tensors
• Efficient Computation of Extreme Excursion Probabilities for Dynamical Systems through Rice's Formula
• A unified performance analysis of likelihood-informed subspace methods
• Derivative-informed projected neural networks for high-dimensional parametric maps governed by PDEs
• The Bayesian inversion problem for thermal average sampling of quantum systems
• A Benchmark for the Bayesian Inversion of Coefficients in Partial Differential Equations
• Geometric MCMC for infinite-dimensional inverse problems
• Optimal design of acoustic metamaterial cloaks under uncertainty
• Object based Bayesian full-waveform inversion for shear elastography
• Multi-physics adjoint modeling of Earth structure: combining gravimetric, seismic, and geodynamic inversions
• Scalable posterior approximations for large-scale Bayesian inverse problems via likelihood-informed parameter and state reduction
• Proposals which speed up function-space MCMC
• An Adaptive Independence Sampler MCMC Algorithm for Bayesian Inferences of Functions
• Sequential implicit sampling methods for Bayesian inverse problems
• Derivative-Free Bayesian Inversion Using Multiscale Dynamics
• Hierarchical Matrix Approximations of Hessians Arising in Inverse Problems Governed by PDEs
• Optimization-Based Markov Chain Monte Carlo Methods for Nonlinear Hierarchical Statistical Inverse Problems
• Efficient Marginalization-Based MCMC Methods for Hierarchical Bayesian Inverse Problems
• Scalable Optimization-Based Sampling on Function Space
• A fast particle-based approach for calibrating a 3-D model of the Antarctic ice sheet
• A stochastic Newton MCMC method for large-scale statistical inverse problems with application to seismic inversion
• Scalable Matrix-Free Adaptive Product-Convolution Approximation for Locally Translation-Invariant Operators
• Deconvolving breath alcohol concentration from biosensor measured transdermal alcohol level under uncertainty: a Bayesian approach
• On an adaptive preconditioned Crank-Nicolson MCMC algorithm for infinite dimensional Bayesian inference
• Variational Bayes' Method for Functions with Applications to Some Inverse Problems
• On global normal linear approximations for nonlinear Bayesian inverse problems
• On the accept-reject mechanism for Metropolis-Hastings algorithms
• PyAlbany: a Python interface to the C++ multiphysics solver Albany
• Joint estimation of Robin coefficient and domain boundary for the Poisson problem
• Optimal experimental design for infinite-dimensional Bayesian inverse problems governed by PDEs: a review
• Multilevel Hierarchical Decomposition of Finite Element White Noise with Application to Multilevel Markov Chain Monte Carlo
• Generalized Computer Model Calibration for Radiation Transport Simulation
• Residual-based error correction for neural operator accelerated Infinite-dimensional Bayesian inverse problems
• CUQIpy: II. Computational uncertainty quantification for PDE-based inverse problems in Python
• Certified coordinate selection for high-dimensional Bayesian inversion with Laplace prior
• Scalable conditional deep inverse Rosenblatt transports using tensor trains and gradient-based dimension reduction
• Algorithm 1040: the Sparse Grids Matlab Kit -- a Matlab implementation of sparse grids for high-dimensional function approximation and uncertainty quantification
• On unifying randomized methods for inverse problems
• Chilled sampling for uncertainty quantification: a motivation from a meteorological inverse problem ^*
• An Offline-Online Decomposition Method for Efficient Linear Bayesian Goal-Oriented Optimal Experimental Design: Application to Optimal Sensor Placement
• Residual-based error corrector operator to enhance accuracy and reliability of neural operator surrogates of nonlinear variational boundary-value problems
• Multilevel hierarchical decomposition of finite element white noise with application to multilevel Markov chain Monte Carlo
• Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems
• Enabling Hyper-Differential Sensitivity Analysis for Ill-Posed Inverse Problems
• Derivative-informed neural operator: an efficient framework for high-dimensional parametric derivative learning
• The Bayesian approach to inverse Robin problems
• Prior normalization for certified likelihood-informed subspace detection of Bayesian inverse problems
• Hyper-differential sensitivity analysis with respect to model discrepancy: optimal solution updating
• Adaptive operator learning for infinite-dimensional Bayesian inverse problems
• Large-scale Bayesian optimal experimental design with derivative-informed projected neural network
• Point spread function approximation of high-rank Hessians with locally supported nonnegative integral kernels
• hIPPYlib-MUQ: a Bayesian inference software framework for integration of data with complex predictive models under uncertainty
• Non-centered parametric variational Bayes’ approach for hierarchical inverse problems of partial differential equations
• Analysis of a Computational Framework for Bayesian Inverse Problems: Ensemble Kalman Updates and MAP Estimators under Mesh Refinement

This page was built for publication: A computational framework for infinite-dimensional Bayesian inverse problems. II: stochastic Newton MCMC with application to ice sheet flow inverse problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2929997)