EUVIMEDEUVIMED entdecken
Crossref

On the Training and Generalization of Deep Operator Networks

Sanghyun Lee, Yeonjong Shin

SIAM Journal on Scientific Computing · 31.08.2024 · 46 (4) · C273-C296

DOI: 10.1137/23m1598751 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SIAM Journal on Scientific Computing, online erschienen am 01.07.2024. Verfasst von Sanghyun Lee und 1 weiteren Autor:innen. Die Arbeit wurde bisher 34-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: National Science Foundation, U.S. Department of Energy, Ministry of Science and ICT of Korea.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a novel training method for deep operator networks (DeepONets), one of the most popular neural network models for operators. DeepONets are constructed by two subnetworks, namely the branch and trunk networks. Typically, the two subnetworks are trained simultaneously, which amounts to solving a complex optimization problem in a high dimensional space. In addition, the nonconvex and nonlinear nature makes training very challenging. To tackle such a challenge, we propose a two-step training method that trains the trunk network first and then sequentially trains the branch network. The core mechanism is motivated by the divide-and-conquer paradigm and is the decomposition of the entire complex training task into two subtasks with reduced complexity. Therein the Gram–Schmidt orthonormalization process is introduced which significantly improves stability and generalization ability. On the theoretical side, we establish a generalization error estimate in terms of the number of training data, the width of DeepONets, and the number of input and output sensors. Numerical examples are presented to demonstrate the effectiveness of the two-step training method, including Darcy flow in heterogeneous porous media.

Förderung

  • National Science Foundation
  • U.S. Department of Energy
  • Ministry of Science and ICT of Korea

Zitieren

APA

Lee, S., & Shin, Y. (2024). On the Training and Generalization of Deep Operator Networks. SIAM Journal on Scientific Computing, 46(4), C273-C296. https://doi.org/10.1137/23m1598751

Vancouver

Lee S, Shin Y. On the Training and Generalization of Deep Operator Networks. SIAM Journal on Scientific Computing. 2024;46(4):C273-C296. doi:10.1137/23m1598751

BibTeX
@article{Lee2024,
  author = {Lee, Sanghyun and Shin, Yeonjong},
  title = {On the Training and Generalization of Deep Operator Networks},
  journal = {SIAM Journal on Scientific Computing},
  year = {2024},
  volume = {46},
  number = {4},
  pages = {C273--C296},
  doi = {10.1137/23m1598751}
}
Kostenlos starten →