English

Learning Hierarchical Priors in VAEs

Machine Learning 2019-10-08 v5 Machine Learning

Abstract

We propose to learn a hierarchical prior in the context of variational autoencoders to avoid the over-regularisation resulting from a standard normal prior distribution. To incentivise an informative latent representation of the data, we formulate the learning problem as a constrained optimisation problem by extending the Taming VAEs framework to two-level hierarchical models. We introduce a graph-based interpolation method, which shows that the topology of the learned latent representation corresponds to the topology of the data manifold---and present several examples, where desired properties of latent representation such as smoothness and simple explanatory factors are learned by the prior.

Keywords

Cite

@article{arxiv.1905.04982,
  title  = {Learning Hierarchical Priors in VAEs},
  author = {Alexej Klushyn and Nutan Chen and Richard Kurle and Botond Cseke and Patrick van der Smagt},
  journal= {arXiv preprint arXiv:1905.04982},
  year   = {2019}
}

Comments

Published at NeurIPS 2019 (spotlight)

R2 v1 2026-06-23T09:04:36.158Z