Dear Spiral user, We’re pleased to introduce Helix, https://helix.imperial.ac.uk/ Imperial’s new repository for research datasets. Please use Helix for all future dataset deposits. There is no change to Spiral - continue depositing as usual. For queries, contact the Research Data Management team at rdm-enquiries@imperial.ac.uk. Kind regards, the Spiral Team. The Spiral team
Learning Green’s functions associated with time-dependent partial differential equations
File(s)
22-0433.pdf (620.67 KB)
Published version
Author(s)
Boullé, N
Kim, S
Shi, T
Townsend, A
Type
Journal Article
Abstract
Neural operators are a popular technique in scientific machine learning to learn a mathematical model of the behavior of unknown physical systems from data. Neural operators are especially useful to learn solution operators associated with partial differential equations (PDEs) from pairs of forcing functions and solutions when numerical solvers are not available or the underlying physics is poorly understood. In this work, we attempt to provide theoretical foundations to understand the amount of training data needed to learn time-dependent PDEs. Given input-output pairs from a parabolic PDE in any spatial dimension n ≥ 1, we derive the first theoretically rigorous scheme for learning the associated solution operator, which takes the form of a convolution with a Green’s function G. Until now, rigorously learning Green’s functions associated with time-dependent PDEs has been a major challenge in the field of scientific machine learning because G may not be square-integrable when n > 1, and time-dependent PDEs have transient dynamics. By combining the hierarchical low-rank structure of G together with randomized numerical linear algebra, we construct an approximant to G that achieves a relative error of O(Γ−1/2) in the L1-norm with high probability by using at most O(− n+2 2 log(1/)) input-output training pairs, where Γ is a measure of the quality of the training dataset for learning G, and > 0 is sufficiently small.
Date Issued
2022-08-01
Date Acceptance
2022-08-01
Citation
Journal of Machine Learning Research, 2022, 23 (218), pp.1-34
URI
ISSN
1532-4435
Publisher
JMLR
Start Page
1
End Page
34
Journal / Book Title
Journal of Machine Learning Research
Volume
23
Issue
218
Copyright Statement
©2022 Nicolas Boull´e, Seick Kim, Tianyi Shi, and Alex Townsend.
License: CC-BY 4.0, see https://creativecommons.org/licenses/by/4.0/. Attribution requirements are provided
at http://jmlr.org/papers/v23/22-0433.html
License URL
Publication Status
Published