ERROR ESTIMATES OF RESIDUAL MINIMIZATION USING NEURAL NETWORKS FOR LINEAR PDES
Author:
Abstract
We propose an abstract framework for analyzing the convergence of least-squares methods based on residual minimization when feasible solutions are neural networks. With the norm relations and compactness arguments, we derive error estimates for both continuous and discrete formulations of residual minimization in strong and weak forms. The formulations cover recently developed physicsinformed neural networks based on strong and variational formulations.
Publisher
Begell House
Subject
General Medicine
Link
https://www.dl.begellhouse.com/download/article/03f9d19a2a5da51f/JMLMC0404(4)-50411.pdf
Reference63 articles.
1. Bai, G., Koley, U., Mishra, S., and Molinaro, R., Physics Informed Neural Networks (PINNs) for Approximating Nonlinear Dispersive PDEs, J. Comput. Math., vol. 39, no. 6, pp. 816-847, 2021.
2. Berg, J. and Nystrom, K., A Unified Deep Artificial Neural Network Approach to Partial Differential Equations in Complex Geometries, Neurocomputing, vol. 317, pp. 28-41, 2018.
3. Bochev, P. and Gunzburger,M., Finite Element Methods of Least-Squares Type, SIAM Rev., vol. 40, no. 4, pp. 789-837, 1998.
4. Bochev, P. and Gunzburger,M., Least-Squares Methods for Hyperbolic Problems, Handbook of Numerical Methods for Hyperbolic Problems Basic and Fundamental Issues, Amsterdam: North Holland, 2016.
5. Bramble, J.H. and Schatz, A.H., Rayleigh-Ritz-GalerkinMethods for Dirichlet's Problem Using Subspaces without Boundary Conditions, Commun. Pure Appl. Math., vol. 23, pp. 653-675, 1970.
Cited by 10 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Moving sampling physics-informed neural networks induced by moving mesh PDE;Neural Networks;2024-12
2. Generalization of PINNs for elliptic interface problems;Applied Mathematics Letters;2024-11
3. Physics-informed deep learning of rate-and-state fault friction;Computer Methods in Applied Mechanics and Engineering;2024-10
4. Inf-sup neural networks for high-dimensional elliptic PDE problems;Journal of Computational Physics;2024-10
5. Physics-informed and data-driven discovery of governing equations for complex phenomena in heterogeneous media;Physical Review E;2024-04-04
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