BCR-Net: a neural network based on the nonstandard wavelet form

Abstract

This paper proposes a novel neural network architecture inspired by the nonstandard form proposed by Beylkin, Coifman, and Rokhlin in [Communications on Pure and Applied Mathematics, 44(2), 141-183]. The nonstandard form is a highly effective wavelet-based compression scheme for linear integral operators. In this work, we first represent the matrix-vector product algorithm of the nonstandard form as a linear neural network where every scale of the multiresolution computation is carried out by a locally connected linear sub-network. In order to address nonlinear problems, we propose an extension, called BCR-Net, by replacing each linear sub-network with a deeper and more powerful nonlinear one. Numerical results demonstrate the efficiency of the new architecture by approximating nonlinear maps that arise in homogenization theory and stochastic computation.

arXiv comment17 pages and 9 figures

References in corpus (12)

• Deep Learning in Neural Networks: An Overview
• A Survey on Deep Learning in Medical Image Analysis
• Sequence to Sequence Learning with Neural Networks
• Solving high-dimensional partial differential equations using deep learning
• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Deep Image Prior
• A unified deep artificial neural network approach to partial differential equations in complex geometries
• Learning Functions: When Is Deep Better Than Shallow
• A multiscale neural network based on hierarchical matrices
• Deep Multiscale Model Learning
• Butterfly-Net: Optimal Function Representation Based on Convolutional Neural Networks
• A multiscale neural network based on hierarchical nested bases

Cited by in corpus (18)

• Fourier Neural Operator for Parametric Partial Differential Equations
• PFNN: A Penalty-Free Neural Network Method for Solving a Class of Second-Order Boundary-Value Problems on Complex Geometries
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• Solving Electrical Impedance Tomography with Deep Learning
• Int-Deep: A Deep Learning Initialized Iterative Method for Nonlinear Problems
• Attention-Enhanced Neural Network Models for Turbulence Simulation
• Symplectic Neural Networks in Taylor Series Form for Hamiltonian Systems
• Variational training of neural network approximations of solution maps for physical models
• Meta-learning Pseudo-differential Operators with Deep Neural Networks
• Parameter Estimation with Dense and Convolutional Neural Networks Applied to the FitzHugh-Nagumo ODE
• Butterfly-Net: Optimal Function Representation Based on Convolutional Neural Networks
• Solving Optical Tomography with Deep Learning
• Optimal Dirichlet Boundary Control by Fourier Neural Operators Applied to Nonlinear Optics
• A reduced order Schwarz method for nonlinear multiscale elliptic equations based on two-layer neural networks
• Solving the inverse problem of microscopy deconvolution with a residual Beylkin-Coifman-Rokhlin neural network
• Butterfly-Net2: Simplified Butterfly-Net and Fourier Transform Initialization
• Learning the geometry of wave-based imaging
• Solver-in-the-Loop joint operator learning: fractional Laplace-Beltrami features for interface reconstruction