Resampled Priors for Variational Autoencoders

Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS) 2019 We propose Learned Accept/Reject Sampling (LARS), a method for constructing richer priors using rejection sampling with a learned acceptance function. This work is motivated by recent analyses o...

Full description

Saved in:
Bibliographic Details
Main Authors Bauer, Matthias, Mnih, Andriy
Format Journal Article
LanguageEnglish
Published 26.10.2018
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS) 2019 We propose Learned Accept/Reject Sampling (LARS), a method for constructing richer priors using rejection sampling with a learned acceptance function. This work is motivated by recent analyses of the VAE objective, which pointed out that commonly used simple priors can lead to underfitting. As the distribution induced by LARS involves an intractable normalizing constant, we show how to estimate it and its gradients efficiently. We demonstrate that LARS priors improve VAE performance on several standard datasets both when they are learned jointly with the rest of the model and when they are fitted to a pretrained model. Finally, we show that LARS can be combined with existing methods for defining flexible priors for an additional boost in performance.
DOI:10.48550/arxiv.1810.11428