Neural Network Element Method for Partial Differential Equations
Yifan Wang, Zhongshuo Lin, Hehu Xie · 2025
Read Neural Network Element Method for Partial Differential Equations by Yifan Wang, Zhongshuo Lin, Hehu Xie (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-16862.txt.
Description
Yifan Wang, Zhongshuo Lin, and Hehu Xie fuse finite elements with neural networks to solve partial differential equations. Classic solvers mesh a domain and solve on grid points; networks fit smoothly but strain at odd shapes. Their element method builds network-based pieces that live on a finite element mesh. Boundary rules then hold exactly, even on complex geometry.
Network fitting power stays high even where solutions spike or kink. The team backs the method with a full error analysis. That theory tells users how accuracy scales with mesh and network size. The aim is engineering problems that broke pure learning solvers. Mesh discipline plus network flexibility is the core trade they offer.
Catalog record
| CREATOR | Yifan Wang, Zhongshuo Lin, Hehu Xie |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-16862