Using and Abusing Equivariance

Conference Paper (2023)
Author(s)

T.F. Edixhoven (Student TU Delft)

Attila Lengyel (TU Delft - Pattern Recognition and Bioinformatics)

J.C. Gemert (TU Delft - Pattern Recognition and Bioinformatics)

Research Group
Pattern Recognition and Bioinformatics
Copyright
© 2023 T.F. Edixhoven, A. Lengyel, J.C. van Gemert
DOI related publication
https://doi.org/10.1109/ICCVW60793.2023.00019
More Info
expand_more
Publication Year
2023
Language
English
Copyright
© 2023 T.F. Edixhoven, A. Lengyel, J.C. van Gemert
Related content
Research Group
Pattern Recognition and Bioinformatics
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. @en
Pages (from-to)
119-128
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

In this paper we show how Group Equivariant Convolutional Neural Networks use subsampling to learn to break equivariance to the rotation and reflection symmetries. We focus on the 2D rotations and reflections and investigate the impact of the broken equivariance on network performance. We show that a change in the input dimension of a network as small as a single pixel can be enough for commonly used architectures to become approximately equivariant, rather than exactly. We investigate the impact of networks not being exactly equivariant and find that approximately equivariant networks generalise significantly worse to unseen symmetries compared to their exactly equivariant counterparts. However, when the symmetries in the training data are not identical to the symmetries of the network, we find that approximately equivariant networks can relax their equivariance constraints, matching or outperforming exactly equivariant networks on common benchmarks.

Files

Edixhoven_Using_and_Abusing_Eq... (pdf)
(pdf | 0 Mb)
- Embargo expired in 30-06-2024
License info not available