CUED Publications database

Training generative neural networks via maximum mean discrepancy optimization

Dziugaite, GK and Roy, DM and Ghahramani, Z (2015) Training generative neural networks via maximum mean discrepancy optimization. In: UNSPECIFIED pp. 258-267..

Full text not available from this repository.

Abstract

We consider training a deep neural network to generate samples from an unknown distribution given i.i.d. data. We frame learning as an optimization minimizing a two-sample test statistic-informally speaking, a good generator network produces samples that cause a twosample test to fail to reject the null hypothesis. As our two-sample test statistic, we use an unbiased estimate of the maximum mean discrepancy, which is the centerpiece of the nonparametric kernel two-sample test proposed by Gretton et al.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Subjects: UNSPECIFIED
Divisions: Div F > Computational and Biological Learning
Depositing User: Cron Job
Date Deposited: 17 Jul 2017 19:36
Last Modified: 03 Aug 2017 03:09
DOI: