HSIC-InfoGAN: Learning Unsupervised Disentangled Representations by Maximising Approximated Mutual Information

Xiao Liu*, Spyridon Thermos, Pedro Sanchez, Alison Q. O’Neil, Sotirios A. Tsaftaris

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Learning disentangled representations requires either supervision or the introduction of specific model designs and learning constraints as biases. InfoGAN is a popular disentanglement framework that learns unsupervised disentangled representations by maximising the mutual information between latent representations and their corresponding generated images. Maximisation of mutual information is achieved by introducing an auxiliary network and training with a latent regression loss. In this short exploratory paper, we study the use of the Hilbert-Schmidt Independence Criterion (HSIC) to approximate mutual information between latent representation and image, termed HSIC-InfoGAN. Directly optimising the HSIC loss avoids the need for an additional auxiliary network. We qualitatively compare the level of disentanglement in each model, suggest a strategy to tune the hyperparameters of HSIC-InfoGAN, and discuss the potential of HSIC-InfoGAN for medical applications.

Original languageEnglish
Title of host publicationMedical Applications with Disentanglements - First MICCAI Workshop, MAD 2022, Held in Conjunction with MICCAI 2022, Proceedings
EditorsJana Fragemann, Jianning Li, Jan Egger, Xiao Liu, Sotirios A. Tsaftaris, Jens Kleesiek
PublisherSpringer
Pages15-21
Number of pages7
ISBN (Print)9783031250453
DOIs
Publication statusPublished - 1 Feb 2023
Event1st MICCAI Workshop on Medical Applications with Disentanglements, MAD 2022, held in conjunction with MICCAI 2022 - Singapore, Singapore
Duration: 22 Sept 202222 Sept 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13823 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference1st MICCAI Workshop on Medical Applications with Disentanglements, MAD 2022, held in conjunction with MICCAI 2022
Country/TerritorySingapore
CitySingapore
Period22/09/2222/09/22

Keywords / Materials (for Non-textual outputs)

  • Disentangled representation learning
  • HSIC
  • InfoGAN

Fingerprint

Dive into the research topics of 'HSIC-InfoGAN: Learning Unsupervised Disentangled Representations by Maximising Approximated Mutual Information'. Together they form a unique fingerprint.

Cite this