Disentangling by Factorising

Abstract

We define and address the problem of unsupervised learning of disentangled representations on data generated from independent factors of variation. We propose FactorVAE, a method that disentangles by encouraging the distribution of representations to be factorial and hence independent across the dimensions. We show that it improves upon ${\hat{I}}^2$-VAE by providing a better trade-off between disentanglement and reconstruction quality. Moreover, we highlight the problems of a commonly used disentanglement metric and introduce a new metric that does not suffer from them.

Shorter version appeared in Learning Disentangled Representations: From Perception to Control workshop at NIPS, 2017: https://sites.google.com/corp/view/disentanglenips2017