Numerical analysis of physics-informed neural networks and related models in physics-informed machine learning
Acta Numerica · 2024-07 · 33 · 633-713
DOI: 10.1017/s0962492923000089 ↗Kurz gefasst
Fachartikel in der Zeitschrift Acta Numerica, online erschienen am 04.09.2024. Verfasst von Tim De Ryck und 1 weiteren Autor:innen. Die Arbeit wurde bisher 89-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Physics-informed neural networks (PINNs) and their variants have been very popular in recent years as algorithms for the numerical simulation of both forward and inverse problems for partial differential equations. This article aims to provide a comprehensive review of currently available results on the numerical analysis of PINNs and related models that constitute the backbone of physics-informed machine learning. We provide a unified framework in which analysis of the various components of the error incurred by PINNs in approximating PDEs can be effectively carried out. We present a detailed review of available results on approximation, generalization and training errors and their behaviour with respect to the type of the PDE and the dimension of the underlying domain. In particular, we elucidate the role of the regularity of the solutions and their stability to perturbations in the error analysis. Numerical results are also presented to illustrate the theory. We identify training errors as a key bottleneck which can adversely affect the overall performance of various models in physics-informed machine learning.
Zitieren
APA
Ryck, T. D., & Mishra, S. (2024). Numerical analysis of physics-informed neural networks and related models in physics-informed machine learning. Acta Numerica, 33, 633-713. https://doi.org/10.1017/s0962492923000089
Vancouver
Ryck TD, Mishra S. Numerical analysis of physics-informed neural networks and related models in physics-informed machine learning. Acta Numerica. 2024;33:633-713. doi:10.1017/s0962492923000089
BibTeX
@article{Ryck2024,
author = {Ryck, Tim De and Mishra, Siddhartha},
title = {Numerical analysis of physics-informed neural networks and related models in physics-informed machine learning},
journal = {Acta Numerica},
year = {2024},
volume = {33},
pages = {633--713},
doi = {10.1017/s0962492923000089}
}
EUVIMED