Least-Squares ReLU Neural Network (LSNN) Method For Linear Advection-Reaction Equation

Abstract

This paper studies least-squares ReLU neural network method for solving the linear advection-reaction problem with discontinuous solution. The method is a discretization of an equivalent least-squares formulation in the set of neural network functions with the ReLU activation function. The method is capable of approximating the discontinuous interface of the underlying problem automatically through the free hyper-planes of the ReLU neural network and, hence, outperforms mesh-based numerical methods in terms of the number of degrees of freedom. Numerical results of some benchmark test problems show that the method can not only approximate the solution with the least number of parameters, but also avoid the common Gibbs phenomena along the discontinuous interface. Moreover, a three-layer ReLU neural network is necessary and sufficient in order to well approximate a discontinuous solution with an interface in ${\mathbb{R}}^2$ that is not a straight line.

arXiv commentsubmitted to Journal of Computational Physics

References in corpus (3)

• DGM: A deep learning algorithm for solving partial differential equations
• Deep least-squares methods: an unsupervised learning-based numerical method for solving elliptic PDEs
• Self-adaptive deep neural network: Numerical approximation to functions and PDEs

Cited by in corpus (10)

• Deep Ritz Method with Adaptive Quadrature for Linear Elasticity
• Neural Control of Discrete Weak Formulations: Galerkin, Least-Squares and Minimal-Residual Methods with Quasi-Optimal Weights
• Self-adaptive deep neural network: Numerical approximation to functions and PDEs
• Cell-average based neural network method for hyperbolic and parabolic partial differential equations
• Least-Squares Neural Network (LSNN) Method For Linear Advection-Reaction Equation: Discontinuity Interface
• An approximate Riemann solver approach in Physics-Informed Neural Networks for hyperbolic conservation laws
• Asymptotic-Preserving Neural Networks for Multiscale Time-Dependent Linear Transport Equations
• Efficient Shallow Ritz Method For 1D Diffusion Problems
• Efficient Shallow Ritz Method For 1D Diffusion-Reaction Problems
• Least-Squares Neural Network (LSNN) Method For Linear Advection-Reaction Equation: Non-constant Jumps