We study the role of latent space dimensionality in Wasserstein auto-encoders (WAEs). Through experimentation on synthetic and real datasets, we argue that random encoders should be preferred over deterministic encoders. We highlight the potential of WAEs for representation learning with promising results on a benchmark disentanglement task.
@article{arxiv.1802.03761,
title = {On the Latent Space of Wasserstein Auto-Encoders},
author = {Paul K. Rubenstein and Bernhard Schoelkopf and Ilya Tolstikhin},
journal= {arXiv preprint arXiv:1802.03761},
year = {2018}
}