Function approximation by deep networks
From MaRDI portal
Publication:2191837
DOI10.3934/cpaa.2020181zbMath1442.62215arXiv1905.12882OpenAlexW3033701124MaRDI QIDQ2191837
Tomaso Poggio, Hrushikesh N. Mhaskar
Publication date: 26 June 2020
Published in: Communications on Pure and Applied Analysis (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/1905.12882
Computational learning theory (68Q32) Applications of graph theory (05C90) Rate of convergence, degree of approximation (41A25) Neural nets and related approaches to inference from stochastic processes (62M45) Statistics on metric spaces (62R20)
Related Items
Neural network approximation ⋮ Feedforward Neural Networks and Compositional Functions with Applications to Dynamical Systems ⋮ A deep network construction that adapts to intrinsic dimensionality beyond the domain ⋮ Approximation of compositional functions with ReLU neural networks ⋮ \textit{hp}-VPINNs: variational physics-informed neural networks with domain decomposition ⋮ Stable recovery of entangled weights: towards robust identification of deep neural networks from minimal samples ⋮ A measure theoretical approach to the mean-field maximum principle for training NeurODEs
Cites Work
- Unnamed Item
- Eignets for function approximation on manifolds
- Function approximation with zonal function networks with activation functions analogous to the rectified linear unit functions
- Approximation properties of a multilayered feedforward artificial neural network
- Limitations of the approximation capabilities of neural networks with one hidden layer
- Error bounds for approximations with deep ReLU networks
- Über die Approximationsordnung bei Kugelfunktionen und algebraischen Polynomen
- Deep vs. shallow networks: An approximation theory perspective
- Localized Linear Polynomial Operators and Quadrature Formulas on the Sphere
- Breaking the Curse of Dimensionality with Convex Neural Networks
