Disentangling Disentanglement in Variational Autoencoders

Emile Mathieu, Tom Rainforth, N Siddharth, Yee Whye Teh

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

Abstract / Description of output

We develop a generalisation of disentanglement in variational autoencoders (VAEs)—decomposition of the latent representation—characterising it as the fulfilment of two factors: a) the latent encodings of the data having an appropriate level of overlap, and b) the aggregate encoding of the data conforming to a desired structure, represented through the prior. Decomposition permits disentanglement, i.e. explicit independence between latents, as a special case, but also allows for a much richer class of properties to be imposed on the learnt representation, such as sparsity, clustering, independent subspaces, or even intricate hierarchical dependency relationships. We show that the β-VAE varies from the standard VAE predominantly in its control of latent overlap and that for the standard choice of an isotropic Gaussian prior, its objective is invariant to rotations of the latent representation. Viewed from the decomposition perspective, breaking this invariance with simple manipulations of the prior can yield better disentanglement with little or no detriment to reconstructions. We further demonstrate how other choices of prior can assist in producing different decompositions and introduce an alternative training objective that allows the control of both decomposition factors in a principled manner.
Original languageEnglish
Title of host publicationProceedings of the 36th International Conference on Machine Learning
EditorsKamalika Chaudhuri, Ruslan Salakhutdinov
Place of PublicationLong Beach, California, USA
PublisherPMLR
Pages4402-4412
Number of pages11
Publication statusPublished - 15 Jun 2019
EventThirty-sixth International Conference on Machine Learning - Long Beach Convention Center, Long Beach, United States
Duration: 9 Jun 201915 Jun 2019
Conference number: 36
https://icml.cc/Conferences/2019

Publication series

NameProceedings of Machine Learning Research
PublisherPMLR
Volume97
ISSN (Electronic)2640-3498

Conference

ConferenceThirty-sixth International Conference on Machine Learning
Abbreviated titleICML 2019
Country/TerritoryUnited States
CityLong Beach
Period9/06/1915/06/19
Internet address

Fingerprint

Dive into the research topics of 'Disentangling Disentanglement in Variational Autoencoders'. Together they form a unique fingerprint.

Cite this