Solving PDEs on Spheres with Physics-Informed Convolutional Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lei, Guanhang, Lei, Zhen, Shi, Lei, Zeng, Chenyu, Zhou, Ding-Xuan
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914898251874304
author Lei, Guanhang
Lei, Zhen
Shi, Lei
Zeng, Chenyu
Zhou, Ding-Xuan
author_facet Lei, Guanhang
Lei, Zhen
Shi, Lei
Zeng, Chenyu
Zhou, Ding-Xuan
contents Physics-informed neural networks (PINNs) have been demonstrated to be efficient in solving partial differential equations (PDEs) from a variety of experimental perspectives. Some recent studies have also proposed PINN algorithms for PDEs on surfaces, including spheres. However, theoretical understanding of the numerical performance of PINNs, especially PINNs on surfaces or manifolds, is still lacking. In this paper, we establish rigorous analysis of the physics-informed convolutional neural network (PICNN) for solving PDEs on the sphere. By using and improving the latest approximation results of deep convolutional neural networks and spherical harmonic analysis, we prove an upper bound for the approximation error with respect to the Sobolev norm. Subsequently, we integrate this with innovative localization complexity analysis to establish fast convergence rates for PICNN. Our theoretical results are also confirmed and supplemented by our experiments. In light of these findings, we explore potential strategies for circumventing the curse of dimensionality that arises when solving high-dimensional PDEs.
format Preprint
id arxiv_https___arxiv_org_abs_2308_09605
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Solving PDEs on Spheres with Physics-Informed Convolutional Neural Networks
Lei, Guanhang
Lei, Zhen
Shi, Lei
Zeng, Chenyu
Zhou, Ding-Xuan
Numerical Analysis
Machine Learning
Statistics Theory
Physics-informed neural networks (PINNs) have been demonstrated to be efficient in solving partial differential equations (PDEs) from a variety of experimental perspectives. Some recent studies have also proposed PINN algorithms for PDEs on surfaces, including spheres. However, theoretical understanding of the numerical performance of PINNs, especially PINNs on surfaces or manifolds, is still lacking. In this paper, we establish rigorous analysis of the physics-informed convolutional neural network (PICNN) for solving PDEs on the sphere. By using and improving the latest approximation results of deep convolutional neural networks and spherical harmonic analysis, we prove an upper bound for the approximation error with respect to the Sobolev norm. Subsequently, we integrate this with innovative localization complexity analysis to establish fast convergence rates for PICNN. Our theoretical results are also confirmed and supplemented by our experiments. In light of these findings, we explore potential strategies for circumventing the curse of dimensionality that arises when solving high-dimensional PDEs.
title Solving PDEs on Spheres with Physics-Informed Convolutional Neural Networks
topic Numerical Analysis
Machine Learning
Statistics Theory
url https://arxiv.org/abs/2308.09605