Variational auto-encoders with Student’s t-prior

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding

Abstract

We propose a new structure for the variational auto-encoders (VAEs) prior, with the weakly informative multivariate Student’s t-distribution. In the proposed model all distribution parameters are trained, thereby allowing for a more robust approximation of the underlying data distribution. We used Fashion-MNIST data in two experiments to compare the proposed VAEs with the standard Gaussian priors. Both experiments showed a better reconstruction of the images with VAEs using Student’s t-prior distribution.

Details

Authors
Organisations
External organisations
  • Halmstad University
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Computer Science
  • Other Computer and Information Science
Original languageEnglish
Title of host publicationESANN 2019 - Proceedings
Subtitle of host publicationThe 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
ISBN (Electronic)978-287-587-065-0
Publication statusPublished - 2019
Publication categoryResearch
Peer-reviewedYes
Event27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning - Bruges, Belgium
Duration: 2019 Apr 242019 Apr 26

Conference

Conference27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Abbreviated titleESANN 2019
CountryBelgium
CityBruges
Period2019/04/242019/04/26

Related projects

View all (1)