Numerical solution of Poisson partial differential equation in high dimension using two-layer neural networks

Author:

Dus Mathias,Ehrlacher Virginie

Abstract

The aim of this article is to analyze numerical schemes using two-layer neural networks with infinite width for the resolution of the high-dimensional Poisson partial differential equation with Neumann boundary condition. Using Barron’s representation of the solution [IEEE Trans. Inform. Theory 39 (1993), pp. 930–945] with a probability measure defined on the set of parameter values, the energy is minimized thanks to a gradient curve dynamic on the 2 2 -Wasserstein space of the set of parameter values defining the neural network. Inspired by the work from Bach and Chizat [On the global convergence of gradient descent for over-parameterized models using optimal transport, 2018; ICM–International Congress of Mathematicians, EMS Press, Berlin, 2023], we prove that if the gradient curve converges, then the represented function is the solution of the elliptic equation considered. Numerical experiments are given to show the potential of the method.

Funder

European Research Council

Publisher

American Mathematical Society (AMS)

Reference24 articles.

1. Universal approximation bounds for superpositions of a sigmoidal function;Barron, Andrew R.;IEEE Trans. Inform. Theory,1993

2. F. Bach and L. Chizat, On the global convergence of gradient descent for over-parameterized models using optimal transport, Adv. Neural Inf. Process. Syst., 2018.

3. Gradient descent on infinitely wide neural networks: global convergence and generalization;Bach, Francis,[2023] \copyright2023

4. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations;Raissi, M.;J. Comput. Phys.,2019

5. Solving high-dimensional partial differential equations using deep learning;Han, Jiequn;Proc. Natl. Acad. Sci. USA,2018

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3