Exploiting Learned Symmetries in Group Equivariant Convolutions

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

Group Equivariant Convolutions (GConvs) enable convolutional neural networks to be equivariant to various transformation groups, but at an additional parameter and compute cost. We investigate the filter parameters learned by GConvs and find certain conditions under which they become highly redundant. We show that GConvs can be efficiently decomposed into depthwise separable convolutions while preserving equivariance properties and demonstrate improved performance and data efficiency on two datasets. All code is publicly available at github.com/Attila94/SepGrouPy.
Original languageEnglish
Title of host publication2021 IEEE International Conference on Image Processing (ICIP)
Subtitle of host publicationProceedings
Place of PublicationPiscataway
PublisherIEEE
Pages759-763
Number of pages5
ISBN (Electronic)978-1-6654-4115-5
ISBN (Print)978-1-6654-3102-6
DOIs
Publication statusPublished - 2021
Event2021 IEEE International Conference on Image Processing (ICIP) - Virtual at Anchorage, United States
Duration: 19 Sep 202122 Sep 2021

Conference

Conference2021 IEEE International Conference on Image Processing (ICIP)
CountryUnited States
CityVirtual at Anchorage
Period19/09/2122/09/21

Keywords

  • group equivariant convolutions
  • depth-wise separable convolutions
  • efficient deep learning

Fingerprint

Dive into the research topics of 'Exploiting Learned Symmetries in Group Equivariant Convolutions'. Together they form a unique fingerprint.

Cite this