English

$\beta$-VAEs can retain label information even at high compression

Machine Learning 2018-12-07 v1 Machine Learning

Abstract

In this paper, we investigate the degree to which the encoding of a β\beta-VAE captures label information across multiple architectures on Binary Static MNIST and Omniglot. Even though they are trained in a completely unsupervised manner, we demonstrate that a β\beta-VAE can retain a large amount of label information, even when asked to learn a highly compressed representation.

Cite

@article{arxiv.1812.02682,
  title  = {$\beta$-VAEs can retain label information even at high compression},
  author = {Emily Fertig and Aryan Arbabi and Alexander A. Alemi},
  journal= {arXiv preprint arXiv:1812.02682},
  year   = {2018}
}

Comments

NeurIPS2018, BDL workshop

R2 v1 2026-06-23T06:34:31.048Z