math.RTcs.LG

Graded Representation Theory of Equivariant Neural Networks

📅 👥 1 authors arXiv:2609.25776v1

Authors

Abstract

Nonlinear activations can create equivariant interactions between irreducible representations that linear maps cannot. We use the Gaussian degree decomposition to extend ordinary polynomial degree to such nonlinear maps, and prove that for a fixed coordinatewise equivariant layer each degree factors into a polynomial determined by the linear maps and a scalar determined by the activation. This separates three distinct obstructions, coming from symmetry, coordinates, and activation.

Paper Information

arXiv ID:
2609.25776v1
Published:

Related Categories

math.RTcs.LG