Multi-scale Deep Neural Network (MscaleDNN) for Solving Poisson-Boltzmann Equation in Complex Domains

Abstract

In this paper, we propose multi-scale deep neural networks (MscaleDNNs) using the idea of radial scaling in frequency domain and activation functions with compact support. The radial scaling converts the problem of approximation of high frequency contents of PDEs' solutions to a problem of learning about lower frequency functions, and the compact support activation functions facilitate the separation of frequency contents of the target function to be approximated by corresponding DNNs. As a result, the MscaleDNNs achieve fast uniform convergence over multiple scales. The proposed MscaleDNNs are shown to be superior to traditional fully connected DNNs and be an effective mesh-less numerical method for Poisson-Boltzmann equations with ample frequency contents over complex and singular domains.

References in corpus (14)

• Adam: A Method for Stochastic Optimization
• Solving high-dimensional partial differential equations using deep learning
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
• Provable approximation properties for deep neural networks
• On the Spectral Bias of Neural Networks
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• The Convergence Rate of Neural Networks for Learned Functions of Different Frequencies
• A mesh-free method for interface problems using the deep learning approach
• Data-Driven, Physics-Based Feature Extraction from Fluid Flow Fields
• Theory of the Frequency Principle for General Deep Neural Networks
• Explicitizing an Implicit Bias of the Frequency Principle in Two-layer Neural Networks
• Multi-scale Deep Neural Networks for Solving High Dimensional PDEs
• Learning to synthesize: splitting and recombining low and high spatial frequencies for image recovery

Cited by in corpus (28)

• Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems
• On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks
• Finite Basis Physics-Informed Neural Networks (FBPINNs): a scalable domain decomposition approach for solving differential equations
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• The Finite Neuron Method and Convergence Analysis
• SPINN: Sparse, Physics-based, and partially Interpretable Neural Networks for PDEs
• A practical PINN framework for multi-scale problems with multi-magnitude loss terms
• Theory of the Frequency Principle for General Deep Neural Networks
• Multi-scale Deep Neural Network (MscaleDNN) Methods for Oscillatory Stokes Flows in Complex Domains
• An augmented Lagrangian deep learning method for variational problems with essential boundary conditions
• MOD-Net: A Machine Learning Approach via Model-Operator-Data Network for Solving PDEs
• Adaptive physics-informed neural operator for coarse-grained non-equilibrium flows
• Solving multiscale elliptic problems by sparse radial basis function neural networks
• Linear Frequency Principle Model to Understand the Absence of Overfitting in Neural Networks
• A shallow physics-informed neural network for solving partial differential equations on surfaces
• Solving Partial Differential Equations with Point Source Based on Physics-Informed Neural Networks
• Reproducing Activation Function for Deep Learning
• Meta-MgNet: Meta Multigrid Networks for Solving Parameterized Partial Differential Equations
• Towards Understanding the Condensation of Neural Networks at Initial Training
• Monte Carlo Neural PDE Solver for Learning PDEs via Probabilistic Representation
• Hierarchical Learning to Solve Partial Differential Equations Using Physics-Informed Neural Networks
• Deep frequency principle towards understanding why deeper learning is faster
• Implicit bias with Ritz-Galerkin method in understanding deep learning for solving PDEs
• A reduced order Schwarz method for nonlinear multiscale elliptic equations based on two-layer neural networks
• Generalization Error Analysis of Neural networks with Gradient Based Regularization
• An Upper Limit of Decaying Rate with Respect to Frequency in Deep Neural Network
• Stationary Density Estimation of Itô Diffusions Using Deep Learning
• Fourier-domain Variational Formulation and Its Well-posedness for Supervised Learning