Skip to main content
Kent Academic Repository

How do Variational Autoencoders Learn? Insights from Representational Similarity

Bonheme, Lisa and Grzes, Marek (2022) How do Variational Autoencoders Learn? Insights from Representational Similarity. [Preprint] (doi:10.48550/arXiv.2205.08399) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided) (KAR id:98783)

The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided.
Official URL:
https://doi.org/10.48550/arXiv.2205.08399

Abstract

The ability of Variational Autoencoders (VAEs) to learn disentangled representations has made them popular for practical applications. However, their behaviour is not yet fully understood. For example, the questions of when they can provide disentangled representations, or suffer from posterior collapse are still areas of active research. Despite this, there are no layerwise comparisons of the representations learned by VAEs, which would further our understanding of these models. In this paper, we thus look into the internal behaviour of VAEs using representational similarity techniques. Specifically, using the CKA and Procrustes similarities, we found that the encoders' representations are learned long before the decoders', and this behaviour is independent of hyperparameters, learning objectives, and datasets. Moreover, the encoders' representations in all but the mean and variance layers are similar across hyperparameters and learning objectives.

Item Type: Preprint
DOI/Identification number: 10.48550/arXiv.2205.08399
Refereed: No
Other identifier: https://arxiv.org/abs/2205.08399
Name of pre-print platform: arXiv
Subjects: Q Science > QA Mathematics (inc Computing science) > QA 76 Software, computer programming,
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
Funders: University of Kent (https://ror.org/00xkeyj56)
Depositing User: Marek Grzes
Date Deposited: 06 Dec 2022 10:58 UTC
Last Modified: 10 Oct 2023 11:25 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/98783 (The current URI for this page, for reference purposes)

University of Kent Author Information

Bonheme, Lisa.

Creator's ORCID:
CReDIT Contributor Roles:

Grzes, Marek.

Creator's ORCID: https://orcid.org/0000-0003-4901-1539
CReDIT Contributor Roles:
  • Depositors only (login required):

Total unique views for this document in KAR since July 2020. For more details click on the image.