Timezone: »

Disentangling by Factorising
Hyunjik Kim · Andriy Mnih

Wed Jul 11 05:50 AM -- 06:00 AM (PDT) @ A7

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 beta-VAE by providing a better trade-off between disentanglement and reconstruction quality and being more robust to the number of training iterations. Moreover, we highlight the problems of a commonly used disentanglement metric and introduce a new metric that does not suffer from them.

Author Information

Hyunjik Kim (DeepMind, University of Oxford)
Andriy Mnih (DeepMind)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors