Learning Symmetrization for Equivariance with Orbit Distance Minimization
Abstract
We present a general framework for symmetrizing an arbitrary neural-network architecture and making it equivariant with respect to a given group. We build upon the proposals of Kim et al. (2023); Kaba et al. (2023) for symmetrization, and improve them by replacing their conversion of neural features into group representations, with an optimization whose loss intuitively measures the distance between group orbits. This change makes our approach applicable to a broader range of matrix groups, such as the Lorentz group O(1, 3), than these two proposals. We experimentally show our method's competitiveness on the SO(2) image classification task, and also its increased generality on the task with O(1, 3). Our implementation will be made accessible at https://github.com/tiendatnguyen-vision/Orbit-symmetrize.
Cite
@article{arxiv.2311.07143,
title = {Learning Symmetrization for Equivariance with Orbit Distance Minimization},
author = {Tien Dat Nguyen and Jinwoo Kim and Hongseok Yang and Seunghoon Hong},
journal= {arXiv preprint arXiv:2311.07143},
year = {2023}
}
Comments
16 pages, 1 figure