Asymptotic-Preserving Neural Networks for Multiscale Time-Dependent Linear Transport Equations

Abstract

In this paper we develop a neural network for the numerical simulation of time-dependent linear transport equations with diffusive scaling and uncertainties. The goal of the network is to resolve the computational challenges of curse-of-dimensionality and multiple scales of the problem. We first show that a standard Physics-Informed Neural Network (PINN) fails to capture the multiscale nature of the problem, hence justifies the need to use Asymptotic-Preserving Neural Networks (APNNs). We show that not all classical AP formulations are fit for the neural network approach. We construct a micro-macro decomposition based neural network, and also build in a mass conservation mechanism into the loss function, in order to capture the dynamic and multiscale nature of the solutions. Numerical examples are used to demonstrate the effectiveness of this APNNs.

References in corpus (10)

• DGM: A deep learning algorithm for solving partial differential equations
• Fourier Neural Operator for Parametric Partial Differential Equations
• Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• An overview on deep learning-based approximation methods for partial differential equations
• Asymptotic-Preserving methods and multiscale models for plasma physics
• Least-Squares ReLU Neural Network (LSNN) Method For Linear Advection-Reaction Equation
• Trend to Equilibrium for the Kinetic Fokker-Planck Equation via the Neural Network Approach
• MIM: A deep mixed residual method for solving high-order partial differential equations
• Multiscale and Nonlocal Learning for PDEs using Densely Connected RNNs