Poster
DVAE#: Discrete Variational Autoencoders with Relaxed Boltzmann Priors
Arash Vahdat · Evgeny Andriyash · William Macready
Room 210 #6
Keywords: [ Variational Inference ] [ Deep Autoencoders ] [ Generative Models ] [ Latent Variable Models ]
Boltzmann machines are powerful distributions that have been shown to be an effective prior over binary latent variables in variational autoencoders (VAEs). However, previous methods for training discrete VAEs have used the evidence lower bound and not the tighter importance-weighted bound. We propose two approaches for relaxing Boltzmann machines to continuous distributions that permit training with importance-weighted bounds. These relaxations are based on generalized overlapping transformations and the Gaussian integral trick. Experiments on the MNIST and OMNIGLOT datasets show that these relaxations outperform previous discrete VAEs with Boltzmann priors. An implementation which reproduces these results is available.
Live content is unavailable. Log in and register to view live content