Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems

DOI10.1016/J.CMA.2017.08.016zbMATH Open1439.65071arXiv1706.06692OpenAlexW2671679770MaRDI QIDQ2310097FDOQ2310097

Authors: Peng Chen, Umberto Villa, Omar Ghattas 

Publication date: 6 April 2020

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Abstract: In this work we propose and analyze a Hessian-based adaptive sparse quadrature to compute infinite-dimensional integrals with respect to the posterior distribution in the context of Bayesian inverse problems with Gaussian prior. Due to the concentration of the posterior distribution in the domain of the prior distribution, a prior-based parametrization and sparse quadrature may fail to capture the posterior distribution and lead to erroneous evaluation results. By using a parametrization based on the Hessian of the negative log-posterior, the adaptive sparse quadrature can effectively allocate the quadrature points according to the posterior distribution. A dimension-independent convergence rate of the proposed method is established under certain assumptions on the Gaussian prior and the integrands. Dimension-independent and faster convergence than $O(N^{-1/2})$ is demonstrated for a linear as well as a nonlinear inverse problem whose posterior distribution can be effectively approximated by a Gaussian distribution at the MAP point.

Full work available at URL: https://arxiv.org/abs/1706.06692

Recommendations

• Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
• Bayes-Hermite quadrature
• Multilevel adaptive sparse Leja approximations for Bayesian inverse problems
• Adaptive Hessian-Based Nonstationary Gaussian Process Response Surface Method for Probability Density Approximation with Application to Bayesian Solution of Large-Scale Inverse Problems
• Scaling limits in computational Bayesian inversion

zbMATH Keywords

curse of dimensionalityGaussian priorsparse griddimension-independent convergence analysisHessian-based adaptive sparse quadratureinfinite-dimensional Bayesian inverse problems

Mathematics Subject Classification ID

Bayesian inference (62F15) Numerical integration (65D30) Numerical solution to inverse problems in abstract spaces (65J22)

Cites Work

• Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
• Riemann manifold Langevin and Hamiltonian Monte Carlo methods. With discussion and authors' reply
• Inverse problems: a Bayesian perspective
• Title not available (Why is that?)
• High-dimensional adaptive sparse polynomial interpolation and applications to parametric PDEs
• Dimension-adaptive tensor-product quadrature
• Convergence rates of best \(N\)-term Galerkin approximations for a class of elliptic SPDEs
• A stochastic Newton MCMC method for large-scale statistical inverse problems with application to seismic inversion
• A computational framework for infinite-dimensional Bayesian inverse problems. I: The linearized case, with application to global seismic inversion
• Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
• Title not available (Why is that?)
• MAP estimators and their consistency in Bayesian nonparametric inverse problems
• An implicit restarted Lanczos method for large symmetric eigenvalue problems
• Sparse deterministic approximation of Bayesian inverse problems
• A computational framework for infinite-dimensional Bayesian inverse problems. II: stochastic Newton MCMC with application to ice sheet flow inverse problems
• Complexity analysis of accelerated MCMC methods for Bayesian inversion
• Dimension-independent likelihood-informed MCMC
• Sparse-grid, reduced-basis Bayesian inversion: nonaffine-parametric nonlinear equations
• Optimal low-rank approximations of Bayesian linear inverse problems
• 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
• An adaptive sparse grid algorithm for elliptic PDEs with lognormal diffusion coefficient
• A new algorithm for high-dimensional uncertainty quantification based on dimension-adaptive sparse grid approximation and reduced basis methods
• Sparse-grid, reduced-basis Bayesian inversion
• Scaling limits in computational Bayesian inversion

Cited In (29)

• Taylor approximation and variance reduction for PDE-constrained optimal control under uncertainty
• Hessian initialization strategies for \(\ell \)-BFGS solving non-linear inverse problems
• Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
• A Fast and Scalable Computational Framework for Large-Scale High-Dimensional Bayesian Optimal Experimental Design
• Non-asymptotic error estimates for the Laplace approximation in Bayesian inverse problems
• Sparse polynomial approximations for affine parametric saddle point problems
• Projected Wasserstein Gradient Descent for High-Dimensional Bayesian Inference
• Adaptive Hessian-Based Nonstationary Gaussian Process Response Surface Method for Probability Density Approximation with Application to Bayesian Solution of Large-Scale Inverse Problems
• Stein variational reduced basis Bayesian inversion
• Quasi-Monte Carlo Bayesian estimation under Besov priors in elliptic inverse problems
• Taylor approximation for chance constrained optimization problems governed by partial differential equations with high-dimensional random parameters
• Efficient estimation of hydraulic conductivity heterogeneity with non-redundant measurement information
• Bayesian inference of heterogeneous epidemic models: application to COVID-19 spread accounting for long-term care facilities
• Learning physics-based models from data: perspectives from inverse problems and model reduction
• Generalized Modes in Bayesian Inverse Problems
• An Offline-Online Decomposition Method for Efficient Linear Bayesian Goal-Oriented Optimal Experimental Design: Application to Optimal Sensor Placement
• Sparse quadrature for high-dimensional integration with Gaussian measure
• How sharp are error bounds? -- Lower bounds on quadrature worst-case errors for analytic functions --
• Derivative-informed neural operator: an efficient framework for high-dimensional parametric derivative learning
• Tensor train construction from tensor actions, with application to compression of large high order derivative tensors
• Derivative-informed projected neural networks for high-dimensional parametric maps governed by PDEs
• Optimal design of acoustic metamaterial cloaks under uncertainty
• On the convergence of the Laplace approximation and noise-level-robustness of Laplace-based Monte Carlo methods for Bayesian inverse problems
• Multilevel adaptive sparse Leja approximations for Bayesian inverse problems
• Large-scale Bayesian optimal experimental design with derivative-informed projected neural network
• Variational inference for nonlinear inverse problems via neural net kernels: comparison to Bayesian neural networks, application to topology optimization
• Variational Bayesian approximation of inverse problems using sparse precision matrices
• Multilevel Quasi-Monte Carlo Uncertainty Quantification for Advection-Diffusion-Reaction
• Finite element discretizations for variable-order fractional diffusion problems

Uses Software

• hIPPYlib

This page was built for publication: Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2310097)