Improved Analysis of PINNs: Alleviate the CoD for Compositional Solutions
From MaRDI portal
Publication:6151354
DOI10.4208/aam.oa-2023-0021MaRDI QIDQ6151354
Pingwen Zhang, Cheng Yuan, Yu Ling Jiao, Jerry Zhijian Yang, Xiliang Lu
Publication date: 11 March 2024
Published in: Annals of Applied Mathematics (Search for Journal in Brave)
Artificial neural networks and deep learning (68T07) Numerical methods for partial differential equations, boundary value problems (65N99)
Cites Work
- Unnamed Item
- Weak adversarial networks for high-dimensional partial differential equations
- Learning constitutive relations from indirect observations using deep neural networks
- The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
- Solving the Kolmogorov PDE by means of deep learning
- Approximation rates for neural networks with encodable weights in smoothness spaces
- The Barron space and the flow-induced function spaces for neural network models
- Learning constitutive relations using symmetric positive definite neural networks
- Nonparametric regression using deep neural networks with ReLU activation function
- Error bounds for approximations with deep ReLU networks
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Universal approximation bounds for superpositions of a sigmoidal function
- A Multivariate Faa di Bruno Formula with Applications
- Feedforward Neural Networks and Compositional Functions with Applications to Dynamical Systems
- Convergence Rate Analysis for Deep Ritz Method
- A Rate of Convergence of Physics Informed Neural Networks for the Linear Second Order Elliptic PDEs
- Deep Network Approximation for Smooth Functions
- Machine Learning and Computational Mathematics
- The Mathematical Theory of Finite Element Methods
- A Regularity Theory for Static Schrödinger Equations on \(\boldsymbol{\mathbb{R}}\)d in Spectral Barron Spaces
