Approximation and Generalization of DeepONets for Learning Operators Arising from a Class of Singularly Perturbed Problems

TL;DR

This paper demonstrates the effectiveness of DeepONets in approximating solutions to one-dimensional singularly perturbed problems, showing promising convergence and generalization properties, especially near boundary layers.

Contribution

First application of DeepONets to singularly perturbed problems, analyzing convergence, generalization, and employing Shishkin meshes for improved approximation.

Findings

Convergence rate of DeepONets is uniform with respect to perturbation parameter.

DeepONets effectively capture boundary layer behavior.

Numerical experiments support the robustness of the method.

Abstract

Singularly perturbed problems present inherent difficulty due to the presence of a thin boundary layer in its solution. To overcome this difficulty, we propose using deep operator networks (DeepONets), a method previously shown to be effective in approximating nonlinear operators between infinite-dimensional Banach spaces. In this paper, we demonstrate for the first time the application of DeepONets to one-dimensional singularly perturbed problems, achieving promising results that suggest their potential as a robust tool for solving this class of problems. We consider the convergence rate of the approximation error incurred by the operator networks in approximating the solution operator, and examine the generalization gap and empirical risk, all of which are shown to converge uniformly with respect to the perturbation parameter. By utilizing Shishkin mesh points as locations of the loss function, we conduct several numerical experiments that provide further support for the effectiveness of operator networks in capturing the singular boundary layer behavior.