Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
From MaRDI portal
Publication:5162358
DOI10.4208/cicp.OA-2020-0085OpenAlexW2914483840MaRDI QIDQ5162358
Yanyang Xiao, Tao Luo, Yaoyu Zhang, Zhi-Qin John Xu, Zheng Ma
Publication date: 2 November 2021
Published in: Communications in Computational Physics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/1901.06523
Related Items (28)
An efficient neural network method with plane wave activation functions for solving Helmholtz equation ⋮ When and why PINNs fail to train: a neural tangent kernel perspective ⋮ On the Exact Computation of Linear Frequency Principle Dynamics and Its Generalization ⋮ Improved architectures and training algorithms for deep operator networks ⋮ High Order Deep Neural Network for Solving High Frequency Partial Differential Equations ⋮ Scientific machine learning through physics-informed neural networks: where we are and what's next ⋮ A Consensus-Based Global Optimization Method with Adaptive Momentum Estimation ⋮ Linearized Learning with Multiscale Deep Neural Networks for Stationary Navier-Stokes Equations with Oscillatory Solutions ⋮ VC-PINN: variable coefficient physics-informed neural network for forward and inverse problems of PDEs with variable coefficient ⋮ SONets: sub-operator learning enhanced neural networks for solving parametric partial differential equations ⋮ MOD-Net: A Machine Learning Approach via Model-Operator-Data Network for Solving PDEs ⋮ Learning high frequency data via the coupled frequency predictor-corrector triangular DNN ⋮ A Hybrid Method for Three-Dimensional Semi-Linear Elliptic Equations ⋮ Unnamed Item ⋮ A conservative hybrid deep learning method for Maxwell-Ampère-Nernst-Planck equations ⋮ Learning Specialized Activation Functions for Physics-Informed Neural Networks ⋮ Physics-informed neural network frameworks for crack simulation based on minimized peridynamic potential energy ⋮ Asymptotic-preserving neural networks for multiscale time-dependent linear transport equations ⋮ Subspace decomposition based DNN algorithm for elliptic type multi-scale PDEs ⋮ Finite basis physics-informed neural networks (FBPINNs): a scalable domain decomposition approach for solving differential equations ⋮ A Neural Network Approach for Homogenization of Multiscale Problems ⋮ On the spectral bias of coupled frequency predictor-corrector triangular DNN: the convergence analysis ⋮ High Order Deep Domain Decomposition Method for Solving High Frequency Interface Problems ⋮ Physical informed neural networks with soft and hard boundary constraints for solving advection-diffusion equations using Fourier expansions ⋮ Multi-scale fusion network: a new deep learning structure for elliptic interface problems ⋮ A composite neural network that learns from multi-fidelity data: application to function approximation and inverse PDE problems ⋮ A non-gradient method for solving elliptic partial differential equations with deep neural networks ⋮ A deep domain decomposition method based on Fourier features
Uses Software
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- Machine learning from a continuous viewpoint. I
- An efficient multigrid strategy for large-scale molecular mechanics optimization
- The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
- A priori estimates of the population risk for two-layer neural networks
- Convergence Estimates for Multigrid Algorithms without Regularity Assumptions
- Relu Deep Neural Networks and Linear Finite Elements
- A Phase Shift Deep Neural Network for High Frequency Approximation and Wave Problems
- Deep Potential: A General Representation of a Many-Body Potential Energy Surface
- Data-Driven, Physics-Based Feature Extraction from Fluid Flow Fields using Convolutional Neural Networks
- Model Reduction with Memory and the Machine Learning of Dynamical Systems
- Uniformly accurate machine learning-based hydrodynamic models for kinetic equations
- A Multiscale Neural Network Based on Hierarchical Matrices
- SwitchNet: A Neural Network Model for Forward and Inverse Scattering Problems
This page was built for publication: Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks