Theoretical Guarantees for the Statistical Finite Element Method

DOI10.1137/21M1463963arXiv2111.07691OpenAlexW3214423602MaRDI QIDQ6188693

Unnamed Author, Andrew B. Duncan, Mark A. Girolami, Jon Cockayne

Publication date: 11 January 2024

Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Abstract: The statistical finite element method (StatFEM) is an emerging probabilistic method that allows observations of a physical system to be synthesised with the numerical solution of a PDE intended to describe it in a coherent statistical framework, to compensate for model error. This work presents a new theoretical analysis of the statistical finite element method demonstrating that it has similar convergence properties to the finite element method on which it is based. Our results constitute a bound on the Wasserstein-2 distance between the ideal prior and posterior and the StatFEM approximation thereof, and show that this distance converges at the same mesh-dependent rate as finite element solutions converge to the true solution. Several numerical examples are presented to demonstrate our theory, including an example which test the robustness of StatFEM when extended to nonlinear quantities of interest.

Full work available at URL: https://arxiv.org/abs/2111.07691

zbMATH Keywords

inverse problemsfinite element methodsWasserstein distanceprobabilistic numericsmodel discrepancy

Mathematics Subject Classification ID

Probabilistic methods, particle methods, etc. for boundary value problems involving PDEs (65N75) PDEs with randomness, stochastic partial differential equations (35R60) Numerical solutions to stochastic differential and integral equations (65C30)

Cites Work

• An Explicit Link between Gaussian Fields and Gaussian Markov Random Fields: The Stochastic Partial Differential Equation Approach
• Bayesian solution uncertainty quantification for differential equations
• The distance between two random vectors wigh given dispersion matrices
• Functional spaces for the theory of elliptic partial differential equations. Transl. from the French by Reinie Erné
• The stochastic finite element method: past, present and future
• On lower bounds for the \(L^ 2\)-Wasserstein metric in a Hilbert space
• The Frechet distance between multivariate normal distributions
• Statistical analysis of differential equations: introducing probability measures on numerical solutions
• The statistical finite element method (statFEM) for coherent synthesis of observation data and model predictions
• Regularity and convergence analysis in Sobolev and Hölder spaces for generalized Whittle-Matérn fields
• Probabilistic solutions to ordinary differential equations as nonlinear Bayesian filtering: a new perspective
• Procrustes metrics on covariance operators and optimal transportation of Gaussian processes
• Numerical methods for stochastic partial differential equations with white noise
• Regularity of the sample paths of a general second order random field
• Inverse problems: A Bayesian perspective
• An Introduction to Computational Stochastic PDEs
• Bayesian inverse problems for functions and applications to fluid mechanics
• Finite Element Representations of Gaussian Processes: Balancing Numerical and Statistical Accuracy
• Stochastic Equations in Infinite Dimensions
• Bayesian Probabilistic Numerical Methods
• Bayesian Numerical Homogenization
• Probabilistic numerics and uncertainty in computations
• Stochastic Partial Differential Equation Based Modelling of Large Space–Time Data Sets
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: Theoretical Guarantees for the Statistical Finite Element Method