Wasserstein Auto-Encoders: Latent Dimensionality and Random Encoders
Lists
Only you see this
Abstract
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.
BibTeX
@inproceedings{rubenstein2018wasserstein,
author = {Paul K. Rubenstein and Bernhard Schölkopf and Ilya O. Tolstikhin},
title = {{Wasserstein Auto-Encoders: Latent Dimensionality and Random Encoders}},
booktitle = {Workshops @ ICLR},
year = {2018},
}