English

ARDO: A Weak Formulation Deep Neural Network Method for Elliptic and Parabolic PDEs Based on Random Differences of Test Functions

Numerical Analysis 2025-09-05 v1 Artificial Intelligence Numerical Analysis

Abstract

We propose ARDO method for solving PDEs and PDE-related problems with deep learning techniques. This method uses a weak adversarial formulation but transfers the random difference operator onto the test function. The main advantage of this framework is that it is fully derivative-free with respect to the solution neural network. This framework is particularly suitable for Fokker-Planck type second-order elliptic and parabolic PDEs.

Keywords

Cite

@article{arxiv.2509.03757,
  title  = {ARDO: A Weak Formulation Deep Neural Network Method for Elliptic and Parabolic PDEs Based on Random Differences of Test Functions},
  author = {Wei Cai and Andrew Qing He},
  journal= {arXiv preprint arXiv:2509.03757},
  year   = {2025}
}
R2 v1 2026-07-01T05:20:07.972Z