CaltechAUTHORS
  A Caltech Library Service

Model Reduction and Neural Networks for Parametric PDEs

Bhattacharya, Kaushik and Hosseini, Bamdad and Kovachki, Nikola B. and Stuart, Andrew M. (2020) Model Reduction and Neural Networks for Parametric PDEs. . (Unpublished) https://resolver.caltech.edu/CaltechAUTHORS:20200527-074228185

[img] PDF - Submitted Version
See Usage Policy.

1979Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200527-074228185

Abstract

We develop a general framework for data-driven approximation of input-output maps between infinite-dimensional spaces. The proposed approach is motivated by the recent successes of neural networks and deep learning, in combination with ideas from model reduction. This combination results in a neural network approximation which, in principle, is defined on infinite-dimensional spaces and, in practice, is robust to the dimension of finite-dimensional approximations of these spaces required for computation. For a class of input-output maps, and suitably chosen probability measures on the inputs, we prove convergence of the proposed approximation methodology. Numerically we demonstrate the effectiveness of the method on a class of parametric elliptic PDE problems, showing convergence and robustness of the approximation scheme with respect to the size of the discretization, and compare our method with existing algorithms from the literature.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
http://arxiv.org/abs/2005.03180arXivDiscussion Paper
ORCID:
AuthorORCID
Bhattacharya, Kaushik0000-0003-2908-5469
Kovachki, Nikola B.0000-0002-3650-2972
Additional Information:Submitted to the editors May 8, 2020. The authors are grateful to Anima Anandkumar, Kamyar Azizzadenesheli, Zongyi Li and Nicholas H. Nelsen for helpful discussions in the general area of neural networks for PDE-defined maps between Hilbert spaces. The work is supported by MEDE-ARL funding (W911NF-12-0022). AMS is also partially supported by NSF (DMS 1818977) and AFOSR (FA9550-17-1-0185). BH is partially supported by a Von Kármán instructorship at the California Institute of Technology.
Funders:
Funding AgencyGrant Number
Army Research LaboratoryW911NF-12-0022
NSFDMS-1818977
Air Force Office of Scientific Research (AFOSR)FA9550-17-1-0185
CaltechUNSPECIFIED
Subject Keywords:approximation theory, deep learning, model reduction, neural networks, partial differential equations
Classification Code:AMS subject classifications: 65N75, 62M45, 68T05, 60H30, 60H15
Record Number:CaltechAUTHORS:20200527-074228185
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200527-074228185
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:103483
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:27 May 2020 15:56
Last Modified:31 Mar 2021 23:33

Repository Staff Only: item control page