Multilevel adaptive sparse Leja approximations for Bayesian inverse problems

DOI10.1137/19M1260293zbMATH Open1432.35250arXiv1904.12204MaRDI QIDQ5216784FDOQ5216784

Authors: Ionuţ-Gabriel Farcaş, Jonas Latz, Tobias Neckel, Hans-Joachim Bungartz, Elisabeth Ullmann 

Publication date: 20 February 2020

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

Abstract: Deterministic interpolation and quadrature methods are often unsuitable to address Bayesian inverse problems depending on computationally expensive forward mathematical models. While interpolation may give precise posterior approximations, deterministic quadrature is usually unable to efficiently investigate an informative and thus concentrated likelihood. This leads to a large number of required expensive evaluations of the mathematical model. To overcome these challenges, we formulate and test a multilevel adaptive sparse Leja algorithm. At each level, adaptive sparse grid interpolation and quadrature are used to approximate the posterior and perform all quadrature operations, respectively. Specifically, our algorithm uses coarse discretizations of the underlying mathematical model to investigate the parameter space and to identify areas of high posterior probability. Adaptive sparse grid algorithms are then used to place points in these areas, and ignore other areas of small posterior probability. The points are weighted Leja points. As the model discretization is coarse, the construction of the sparse grid is computationally efficient. On this sparse grid, the posterior measure can be approximated accurately with few expensive, fine model discretizations. The efficiency of the algorithm can be enhanced further by exploiting more than two discretization levels. We apply the proposed multilevel adaptive sparse Leja algorithm in numerical experiments involving elliptic inverse problems in 2D and 3D space, in which we compare it with Markov chain Monte Carlo sampling and a standard multilevel approximation.

Full work available at URL: https://arxiv.org/abs/1904.12204

Recommendations

• Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
• Multilevel higher-order quasi-Monte Carlo Bayesian estimation
• Adaptive sparse grid model order reduction for fast Bayesian estimation and inversion
• Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems
• Adaptive construction of surrogates for the Bayesian solution of inverse problems

zbMATH Keywords

Bayesian inferencepartial differential equationmultilevel methodadaptive sparse grids

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Bayesian inference (62F15) Learning and adaptive systems in artificial intelligence (68T05) Numerical integration (65D30) Boundary value problems for second-order elliptic equations (35J25) Inverse problems for PDEs (35R30) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30)

Cites Work

• Fast sampling of parameterised Gaussian random fields
• Sequential Monte Carlo Samplers
• Bayesian calibration of computer models. (With discussion)
• Statistical and computational inverse problems.
• Handbook of Markov Chain Monte Carlo
• Sparse grids
• Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates
• Inverse problems: a Bayesian perspective
• Parameter and state model reduction for large-scale statistical inverse problems
• Title not available (Why is that?)
• Dimension-adaptive tensor-product quadrature
• Inverse Problem Theory and Methods for Model Parameter Estimation
• Adaptive Leja sparse grid constructions for stochastic collocation and high-dimensional approximation
• Title not available (Why is that?)
• An efficient Bayesian inference approach to inverse problems based on an adaptive sparse grid collocation method
• A stochastic collocation approach to Bayesian inference in inverse problems
• Title not available (Why is that?)
• Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
• Adaptive Smolyak Pseudospectral Approximations
• Adaptive multi-fidelity polynomial chaos approach to Bayesian inference in inverse problems
• Adaptive construction of surrogates for the Bayesian solution of inverse problems
• Sparse deterministic approximation of Bayesian inverse problems
• Sparsity in Bayesian inversion of parametric operator equations
• A Hierarchical Multilevel Markov Chain Monte Carlo Algorithm with Applications to Uncertainty Quantification in Subsurface Flow
• On the Lebesgue constant of weighted Leja points for Lagrange interpolation on unbounded domains
• On the Well-posedness of Bayesian Inverse Problems
• Multilevel sequential Monte Carlo for Bayesian inverse problems
• Sparse pseudo spectral projection methods with directional adaptation for uncertainty quantification
• Survey of multifidelity methods in uncertainty propagation, inference, and optimization
• Sampling-free Bayesian inversion with adaptive hierarchical tensor representations
• Sensitivity-driven adaptive sparse stochastic approximations in plasma microinstability analysis
• A transport-based multifidelity preconditioner for Markov chain Monte Carlo
• Adaptive sparse grid model order reduction for fast Bayesian estimation and inversion
• Posterior consistency for Gaussian process approximations of Bayesian posterior distributions
• Convergence analysis of surrogate-based methods for Bayesian inverse problems
• Hessian-based adaptive sparse quadrature for infinite-dimensional Bayesian inverse problems
• Multilevel higher-order quasi-Monte Carlo Bayesian estimation
• Multilevel adaptive stochastic collocation with dimensionality reduction
• Smolyak's algorithm: a powerful black box for the acceleration of scientific computations
• Inverse problems. Tikhonov theory and algorithms
• Goal-oriented adaptive surrogate construction for stochastic inversion
• Sparse-grid, reduced-basis Bayesian inversion
• Bayesian inference using intermediate distribution based on coarse multiscale model for time fractional diffusion equations
• Quasi-Monte Carlo and multilevel Monte Carlo methods for computing posterior expectations in elliptic inverse problems
• On expansions and nodes for sparse grid collocation of lognormal elliptic PDEs

Cited In (11)

• Sparse, adaptive Smolyak quadratures for Bayesian inverse problems
• Stochastic multi-fidelity surrogate modeling of dendritic crystal growth
• Practical considerations when using sparse grids with Bayesian inference for parameter estimation
• Adaptive Leja sparse grid constructions for stochastic collocation and high-dimensional approximation
• An hp‐adaptive multi‐element stochastic collocation method for surrogate modeling with information re‐use
• Multi-fidelity Bayesian optimization to solve the inverse Stefan problem
• Stein Variational Reduced Basis Bayesian Inversion
• Calibrate, emulate, sample
• Variational inference for nonlinear inverse problems via neural net kernels: comparison to Bayesian neural networks, application to topology optimization
• On expansions and nodes for sparse grid collocation of lognormal elliptic PDEs
• Adaptive sparse grid model order reduction for fast Bayesian estimation and inversion

This page was built for publication: Multilevel adaptive sparse Leja approximations for Bayesian inverse problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5216784)