Skip to yearly menu bar Skip to main content


Poster

Controlling Posterior Collapse by an Inverse Lipschitz Constraint on the Decoder Network

Yuri Kinoshita · Kenta Oono · Kenji Fukumizu · Yuichi Yoshida · Shin-ichi Maeda

Exhibit Hall 1 #212
[ ]
[ PDF [ Poster

Abstract:

Variational autoencoders (VAEs) are one of the deep generative models that have experienced enormous success over the past decades. However, in practice, they suffer from a problem called posterior collapse, which occurs when the posterior distribution coincides, or collapses, with the prior taking no information from the latent structure of the input data into consideration. In this work, we introduce an inverse Lipschitz neural network into the decoder and, based on this architecture, provide a new method that can control in a simple and clear manner the degree of posterior collapse for a wide range of VAE models equipped with a concrete theoretical guarantee. We also illustrate the effectiveness of our method through several numerical experiments.

Chat is not available.