A multi-resolution, non-parametric, Bayesian framework for identification of spatially-varying model parameters

DOI10.1016/J.JCP.2009.05.016zbMATH Open1190.62211arXiv0810.0744OpenAlexW2125123544MaRDI QIDQ843476FDOQ843476

Authors: Phaedon-Stelios Koutsourelakis 

Publication date: 12 October 2009

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

Abstract: This paper proposes a hierarchical, multi-resolution framework for the identification of model parameters and their spatially variability from noisy measurements of the response or output. Such parameters are frequently encountered in PDE-based models and correspond to quantities such as density or pressure fields, elasto-plastic moduli and internal variables in solid mechanics, conductivity fields in heat diffusion problems, permeability fields in fluid flow through porous media etc. The proposed model has all the advantages of traditional Bayesian formulations such as the ability to produce measures of confidence for the inferences made and providing not only predictive estimates but also quantitative measures of the predictive uncertainty. In contrast to existing approaches it utilizes a parsimonious, non-parametric formulation that favors sparse representations and whose complexity can be determined from the data. The proposed framework in non-intrusive and makes use of a sequence of forward solvers operating at various resolutions. As a result, inexpensive, coarse solvers are used to identify the most salient features of the unknown field(s) which are subsequently enriched by invoking solvers operating at finer resolutions. This leads to significant computational savings particularly in problems involving computationally demanding forward models but also improvements in accuracy. It is based on a novel, adaptive scheme based on Sequential Monte Carlo sampling which is embarrassingly parallelizable and circumvents issues with slow mixing encountered in Markov Chain Monte Carlo schemes.

Full work available at URL: https://arxiv.org/abs/0810.0744

Recommendations

• Reduced-order model tracking and interpolation to solve PDE-based Bayesian inverse problems
• A Bayesian approach to multiscale inverse problems using the sequential Monte Carlo method
• A Bayesian approach to multiscale inverse problems with on-the-fly scale determination
• Multilevel sequential Monte Carlo for Bayesian inverse problems
• Sparse variational Bayesian approximations for nonlinear inverse problems: applications in nonlinear elastography

zbMATH Keywords

BayesianMonte Carlouncertaintystatistical learningcomplex systems

Mathematics Subject Classification ID

Bayesian inference (62F15) Monte Carlo methods (65C05) Applications of statistics to physics (62P35) Nonparametric inference (62G99)

Cites Work

• Sequential Monte Carlo Methods in Practice
• Sequential Monte Carlo Samplers
• Reversible jump Markov chain Monte Carlo computation and Bayesian model determination
• Bayesian calibration of computer models. (With discussion)
• Sequential importance sampling for nonparametric Bayes models: The next generation
• Title not available (Why is that?)
• Statistical and computational inverse problems.
• Central limit theorem for sequential Monte Carlo methods and its application to Bayesian inference
• Title not available (Why is that?)
• 10.1162/15324430152748236
• Title not available (Why is that?)
• A Correspondence Between Bayesian Estimation on Stochastic Processes and Smoothing by Splines
• Monte Carlo strategies in scientific computing
• Random heterogeneous materials. Microstructure and macroscopic properties
• Generalised Gibbs sampler and multigrid Monte Carlo for Bayesian computation
• Equation-free, coarse-grained multiscale computation: enabling microscopic simulators to perform system-level analysis
• A posteriori error estimation in finite element analysis
• Computational Methods for Inverse Problems
• Title not available (Why is that?)
• A multiscale finite element method for elliptic problems in composite materials and porous media
• Space and space-time modeling using process convolutions
• Computational inelasticity
• The variational multiscale method -- a paradigm for computational mechanics
• Tikhonov regularization of large linear problems
• Bayesian Forecasting for Complex Systems Using Computer Simulators
• A Bayesian calibration approach to the thermal problem
• Atomic decomposition by basis pursuit
• Inverse Problems and Torricelli's Law
• The heterogeneous multiscale methods
• Coarse-gradient Langevin algorithms for dynamic data integration and uncertainty quantification
• Capturing Small Scales in Elliptic Problems Using a Residual-Free Bubbles Finite Element Method
• Hierarchical Bayesian models for inverse problems in heat conduction
• A Bayesian approach to characterizing uncertainty in inverse problems using coarse and fine-scale information
• A stochastic variational multiscale method for diffusion in heterogeneous random media
• A Bayesian inference approach to the inverse heat conduction problem
• Wavelet-Based Numerical Homogenization
• Stochastic upscaling in solid mechanics: an excercise in machine learning
• Multiscale modeling. A Bayesian perspective
• Reversible Jump Markov Chain Monte Carlo Strategies for Bayesian Model Selection in Autoregressive Processes
• A Bayesian analysis of the thermal challenge problem
• A Markov random field model of contamination source identification in porous media flow
• Uncertainties in parameter estimation: The inverse problem
• Uncertainties in parameter estimation: The optimal experiment design
• Fully Bayesian Reconstructions From Single-Photon Emission Computed Tomography Data
• Bayesian Inversion of Geoelectrical Resistivity Data
• An application of Bayesian inverse methods to vertical deconvolution of hydraulic conductivity in a heterogeneous aquifer at Oak Ridge National Laboratory

Cited In (27)

• Multi-fidelity classification using Gaussian processes: accelerating the prediction of large-scale computational models
• Sparse variational Bayesian approximations for nonlinear inverse problems: applications in nonlinear elastography
• Stochastic data assimilation of the random shallow water model loads with uncertain experimental measurements
• A moment-matching method to study the variability of phenomena described by partial differential equations
• Hierarchical sparse observation models and informative prior for Bayesian inference of spatially varying parameters
• Multilevel sequential Monte Carlo for Bayesian inverse problems
• Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
• A Bayesian approach to multiscale inverse problems with on-the-fly scale determination
• Model-based structural health monitoring of naval ship hulls
• Reduced-order model tracking and interpolation to solve PDE-based Bayesian inverse problems
• Stochastic identification of imperfections in a submerged shell structure
• Intercorrelated random fields with bounds and the Bayesian identification of their parameters: Application to linear elastic struts and fibers
• An adaptive sparse-grid iterative ensemble Kalman filter approach for parameter field estimation
• Multi-agent reinforcement learning aided sampling algorithms for a class of multiscale inverse problems
• Parameter identification and uncertainty quantification in stochastic state space models and its application to texture analysis
• Semi-supervised invertible neural operators for Bayesian inverse problems
• A probabilistic graphical model approach to stochastic multiscale partial differential equations
• Selection of model discrepancy priors in Bayesian calibration
• A variational Bayesian method to inverse problems with impulsive noise
• Bayesian methods for characterizing unknown parameters of material models
• Multifidelity approaches for uncertainty quantification
• Multilevel sequential importance sampling for rare event estimation
• Kernel-based approximation methods for partial differential equations: deterministic or stochastic problems?
• Certified Dimension Reduction for Bayesian Updating with the Cross-Entropy Method
• A Bayesian framework for constitutive model identification via use of full field measurements, with application to heterogeneous materials
• Weak neural variational inference for solving Bayesian inverse problems \textit{without} forward models: applications in elastography
• Multimodal, high-dimensional, model-based, Bayesian inverse problems with applications in biomechanics

Uses Software

• PDCO
• BayesDA

This page was built for publication: A multi-resolution, non-parametric, Bayesian framework for identification of spatially-varying model parameters

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q843476)