$E(2)$-Equivariant Vision Transformer
Renjun Xu, Kaifan Yang, Ke Liu, Fengxiang He
Abstract
Vision Transformer (ViT) has achieved remarkable performance in computer vision. However, positional encoding in ViT makes it substantially difficult to learn the intrinsic equivariance in data. Ini- tial attempts have been made on designing equiv- ariant ViT but are proved defective in some cases in this paper. To address this issue, we design a Group Equivariant Vision Transformer (GE-ViT) via a novel, effective positional encoding opera- tor. We prove that GE-ViT meets all the theoreti- cal requirements of an equivariant neural network. Comprehensive experiments are conducted on standard benchmark datasets, demonstrating that GE-ViT significantly outperforms non-equivariant self-attention networks. The code is available at https://github.com/ZJUCDSYangKaifan/GEVit.
BibTeX
@InProceedings{pmlr-v216-xu23b,
title = {$E(2)$-Equivariant Vision Transformer},
author = {Xu, Renjun and Yang, Kaifan and Liu, Ke and He, Fengxiang},
booktitle = {Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence},
pages = {2356--2366},
year = {2023},
editor = {Evans, Robin J. and Shpitser, Ilya},
volume = {216},
series = {Proceedings of Machine Learning Research},
month = {31 Jul--04 Aug},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v216/xu23b/xu23b.pdf},
url = {https://proceedings.mlr.press/v216/xu23b.html},
abstract = {Vision Transformer (ViT) has achieved remarkable performance in computer vision. However, positional encoding in ViT makes it substantially difficult to learn the intrinsic equivariance in data. Ini- tial attempts have been made on designing equiv- ariant ViT but are proved defective in some cases in this paper. To address this issue, we design a Group Equivariant Vision Transformer (GE-ViT) via a novel, effective positional encoding opera- tor. We prove that GE-ViT meets all the theoreti- cal requirements of an equivariant neural network. Comprehensive experiments are conducted on standard benchmark datasets, demonstrating that GE-ViT significantly outperforms non-equivariant self-attention networks. The code is available at https://github.com/ZJUCDSYangKaifan/GEVit.}
}