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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cai, Wei, He, Andrew Qing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918135320281088
author Cai, Wei
He, Andrew Qing
author_facet Cai, Wei
He, Andrew Qing
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2509_03757
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ARDO: A Weak Formulation Deep Neural Network Method for Elliptic and Parabolic PDEs Based on Random Differences of Test Functions
Cai, Wei
He, Andrew Qing
Numerical Analysis
Artificial Intelligence
35Q68, 65N99, 68T07, 76M99
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.
title ARDO: A Weak Formulation Deep Neural Network Method for Elliptic and Parabolic PDEs Based on Random Differences of Test Functions
topic Numerical Analysis
Artificial Intelligence
35Q68, 65N99, 68T07, 76M99
url https://arxiv.org/abs/2509.03757