A Least-Squares-Based Neural Network (LS-Net) for Solving Linear Parametric PDEs
Abstract
Developing efficient methods for solving parametric partial differential equations is crucial for addressing inverse problems. This work introduces a Least-Squares-based Neural Network (LS-Net) method for solving linear parametric PDEs. It utilizes a separated representation form for the parametric PDE solution via a deep neural network and a least-squares solver. In this approach, the output of the deep neural network consists of a vector-valued function, interpreted as basis functions for the parametric solution space, and the least-squares solver determines the optimal solution within the constructed solution space for each given parameter. The LS-Net method requires a quadratic loss function for the least-squares solver to find optimal solutions given the set of basis functions. In this study, we consider loss functions derived from the Deep Fourier Residual and Physics-Informed Neural Networks approaches. We also provide theoretical results similar to the Universal Approximation Theorem, stating that there exists a sufficiently large neural network that can theoretically approximate solutions of parametric PDEs with the desired accuracy. We illustrate the LS-net method by solving one- and two-dimensional problems. Numerical results clearly demonstrate the method's ability to approximate parametric solutions.
Cite
@article{arxiv.2410.15089,
title = {A Least-Squares-Based Neural Network (LS-Net) for Solving Linear Parametric PDEs},
author = {Shima Baharlouei and Jamie M. Taylor and Carlos Uriarte and David Pardo},
journal= {arXiv preprint arXiv:2410.15089},
year = {2025}
}
Comments
It is very important to mention all fund sources, specifically this one: the Marie Sklodowska-Curie grant agreement No 101119556 (IN-DEEP)