Approximation properties of deep ReLU CNNs

Juncai He, Lin Li, Jinchao Xu

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

This paper focuses on establishing L2 approximation properties for deep ReLU convolutional neural networks (CNNs) in two-dimensional space. The analysis is based on a decomposition theorem for convolutional kernels with a large spatial size and multi-channels. Given the decomposition result, the property of the ReLU activation function, and a specific structure for channels, a universal approximation theorem of deep ReLU CNNs with classic structure is obtained by showing its connection with one-hidden-layer ReLU neural networks (NNs). Furthermore, approximation properties are obtained for one version of neural networks with ResNet, pre-act ResNet, and MgNet architecture based on connections between these networks.

Original languageEnglish (US)
Article number38
JournalResearch in Mathematical Sciences
Volume9
Issue number3
DOIs
StatePublished - Sep 2022

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Mathematics (miscellaneous)
  • Computational Mathematics
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Approximation properties of deep ReLU CNNs'. Together they form a unique fingerprint.

Cite this