Universal Approximation Theorem for Equivariant Maps by Group CNNs

Wataru Kumagai, Akiyoshi Sannai

Keywords: Universal Approximation Theorem, CNN, Deep Learning, Symmetry

Abstract: Group symmetry is inherent in a wide variety of data distributions. Data processing that preserves symmetry is described as an equivariant map and often effective in achieving high performance. Convolutional neural networks (CNNs) have been known as models with equivariance and shown to approximate equivariant maps for some specific groups. However, universal approximation theorems for CNNs have been separately derived with individual techniques according to each group and setting. This paper provides a unified method to obtain universal approximation theorems for equivariant maps by CNNs in various settings. As its significant advantage, we can handle non-linear equivariant maps between infinite-dimensional spaces for non-compact groups.

One-sentence Summary: This paper provides a unified method to obtain universal approximation theorems for equivariant maps by CNNs in various settings.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=6nUMpOtPrA