Browse by author
Lookup NU author(s): Dr Junyang Wang, Professor Chris Oates
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
© 2021, The Author(s). The numerical solution of differential equations can be formulated as an inference problem to which formal statistical approaches can be applied. However, nonlinear partial differential equations (PDEs) pose substantial challenges from an inferential perspective, most notably the absence of explicit conditioning formula. This paper extends earlier work on linear PDEs to a general class of initial value problems specified by nonlinear PDEs, motivated by problems for which evaluations of the right-hand-side, initial conditions, or boundary conditions of the PDE have a high computational cost. The proposed method can be viewed as exact Bayesian inference under an approximate likelihood, which is based on discretisation of the nonlinear differential operator. Proof-of-concept experimental results demonstrate that meaningful probabilistic uncertainty quantification for the unknown solution of the PDE can be performed, while controlling the number of times the right-hand-side, initial and boundary conditions are evaluated. A suitable prior model for the solution of PDEs is identified using novel theoretical analysis of the sample path properties of Matérn processes, which may be of independent interest.
Author(s): Wang J, Cockayne J, Chkrebtii O, Sullivan TJ, Oates CJ
Publication type: Article
Publication status: Published
Journal: Statistics and Computing
Year: 2021
Volume: 31
Issue: 5
Online publication date: 27/07/2021
Acceptance date: 11/07/2021
Date deposited: 17/08/2021
ISSN (print): 0960-3174
ISSN (electronic): 1573-1375
Publisher: Springer Nature
URL: https://doi.org/10.1007/s11222-021-10030-w
DOI: 10.1007/s11222-021-10030-w
Altmetrics provided by Altmetric