Exponential Convergence of the Deep Neural Network Approximation for Analytic Functions

Abstract

We prove that for analytic functions in low dimension, the convergence rate of the deep neural network approximation is exponential.

References in corpus (2)

• The Expressive Power of Neural Networks: A View from the Width
• Benefits of depth in neural networks

Cited by in corpus (14)

• PDE-Net 2.0: Learning PDEs from Data with A Numeric-Symbolic Hybrid Deep Network
• Deep Network Approximation for Smooth Functions
• Neural Network Approximation: Three Hidden Layers Are Enough
• Nonlinear Approximation and (Deep) ReLU Networks
• Nonlinear Approximation via Compositions
• Deep Network with Approximation Error Being Reciprocal of Width to Power of Square Root of Depth
• Two-Layer Neural Networks for Partial Differential Equations: Optimization and Generalization Theory
• Full error analysis for the training of deep neural networks
• Structure Probing Neural Network Deflation
• The gap between theory and practice in function approximation with deep neural networks
• Reproducing Activation Function for Deep Learning
• Selection dynamics for deep neural networks
• Deep Neural Networks Are Effective At Learning High-Dimensional Hilbert-Valued Functions From Limited Data
• Analytic Continuation of Noisy Data Using Adams Bashforth ResNet