Transformation-Equivariant Representation Learning with Barber-Agakov and InfoNCE Mutual Information Estimation

Marshal Arijona Sinaga, T. Basarrudin, Adila Alfa Krisnadhi

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The success of deep learning on computer vision tasks is due to the convolution layer being equivariant to the translation. Several works attempt to extend the notion of equivariance into more general transformations. Autoencoding variational transformation (AVT) achieves state of art by approaching the problem from the information theory perspective. The model involves the computation of mutual information, which leads to a more general transformation-equivariant representation model. In this research, we investigate the alternatives of AVT called variational transformation-equivariant (VTE). We utilize the Barber-Agakov and information noise contrastive mutual information estimation to optimize VTE. Furthermore, we also propose a sequential mechanism that involves a self-supervised learning model called predictive-transformation to train our VTE. Results of experiments demonstrate that VTE outperforms AVT on image classification tasks.

Original languageEnglish
Title of host publicationICPRAM 2022 - Proceedings of the 11th International Conference on Pattern Recognition Applications and Methods, Volume 1
EditorsMaria De Marsico, Gabriella Sanniti di Baja, Ana L.N. Fred
PublisherScience and Technology Publications, Lda
Pages99-109
Number of pages11
ISBN (Print)9789897585494
DOIs
Publication statusPublished - 2022
Event11th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2022 - Virtual, Online
Duration: 3 Feb 20225 Feb 2022

Publication series

NameInternational Conference on Pattern Recognition Applications and Methods
Volume1
ISSN (Electronic)2184-4313

Conference

Conference11th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2022
CityVirtual, Online
Period3/02/225/02/22

Keywords

  • Barber-Agakov
  • InfoNCE
  • Mutual Information Estimation
  • Representation Learning
  • Transformation-Equivariant

Fingerprint

Dive into the research topics of 'Transformation-Equivariant Representation Learning with Barber-Agakov and InfoNCE Mutual Information Estimation'. Together they form a unique fingerprint.

Cite this