How Jellyfish Characterise Alternating Group Equivariant Neural Networks

Edward Pearce-Crump

Abstract

We provide a full characterisation of all of the possible alternating group ($A_n$) equivariant neural networks whose layers are some tensor power of $\mathbb{R}^{n}$. In particular, we find a basis of matrices for the learnable, linear, $A_n$-equivariant layer functions between such tensor power spaces in the standard basis of $\mathbb{R}^{n}$. We also describe how our approach generalises to the construction of neural networks that are equivariant to local symmetries.
BibTeX
@inproceedings{pearcecrump2023how,
  author = {Edward Pearce-Crump},
  title = {{How Jellyfish Characterise Alternating Group Equivariant Neural Networks}},
  booktitle = {International Conference on Machine Learning},
  year = {2023},
  url = {https://openreview.net/forum?id=MORCsaQjR7},
}
Figure