pith. sign in

arxiv: 1803.04357 · v2 · pith:BSKERTVAnew · submitted 2018-03-12 · 💻 cs.LG · cs.NE

Learning the Base Distribution in Implicit Generative Models

classification 💻 cs.LG cs.NE
keywords learninggenerativemodelcomplicateddatadistributiondistributionsgans
0
0 comments X
read the original abstract

Popular generative model learning methods such as Generative Adversarial Networks (GANs), and Variational Autoencoders (VAE) enforce the latent representation to follow simple distributions such as isotropic Gaussian. In this paper, we argue that learning a complicated distribution over the latent space of an auto-encoder enables more accurate modeling of complicated data distributions. Based on this observation, we propose a two stage optimization procedure which maximizes an approximate implicit density model. We experimentally verify that our method outperforms GANs and VAEs on two image datasets (MNIST, CELEB-A). We also show that our approach is amenable to learning generative model for sequential data, by learning to generate speech and music.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.