Skip to yearly menu bar Skip to main content


Poster

DVAE++: Discrete Variational Autoencoders with Overlapping Transformations

Arash Vahdat · William Macready · Zhengbing Bian · Amir Khoshaman · Evgeny Andriyash

Hall B #85

Abstract:

Training of discrete latent variable models remains challenging because passing gradient information through discrete units is difficult. We propose a new class of smoothing transformations based on a mixture of two overlapping distributions, and show that the proposed transformation can be used for training binary latent models with either directed or undirected priors. We derive a new variational bound to efficiently train with Boltzmann machine priors. Using this bound, we develop DVAE++, a generative model with a global discrete prior and a hierarchy of convolutional continuous variables. Experiments on several benchmarks show that overlapping transformations outperform other recent continuous relaxations of discrete latent variables including Gumbel-Softmax (Maddison et al., 2016; Jang et al., 2016), and discrete variational autoencoders (Rolfe 2016).

Live content is unavailable. Log in and register to view live content