CUED Publications database

GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution

Kusner, MJ and Hernández-Lobato, JM GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution. (Unpublished)

Full text not available from this repository.

Abstract

Generative Adversarial Networks (GAN) have limitations when the goal is to generate sequences of discrete elements. The reason for this is that samples from a distribution on discrete objects such as the multinomial are not differentiable with respect to the distribution parameters. This problem can be avoided by using the Gumbel-softmax distribution, which is a continuous approximation to a multinomial distribution parameterized in terms of the softmax function. In this work, we evaluate the performance of GANs based on recurrent neural networks with Gumbel-softmax output distributions in the task of generating sequences of discrete elements.

Item Type: Article
Uncontrolled Keywords: stat.ML stat.ML cs.LG
Subjects: UNSPECIFIED
Divisions: Div F > Computational and Biological Learning
Depositing User: Cron Job
Date Deposited: 17 Jul 2017 20:07
Last Modified: 18 Jul 2017 08:51
DOI: