#machine learning
Mar 2026
The Geometry of Polynomial Group Convolutional Neural Networks
A new mathematical framework for PGCNNs using the language of graded group algebras is introduced, which yields two natural parameterizations of the architecture, based on Hadamard and Kronecker products, related by a linear map.
Yacoub Hendi, Daniel Persson, Magdalena Larfors
· arXiv.org · 3 citations