ALICE: Towards understanding adversarial learning for joint distribution matching

Published

Conference Paper

© 2017 Neural information processing systems foundation. All rights reserved. We investigate the non-identifiability issues associated with bidirectional adversarial training for joint distribution matching. Within a framework of conditional entropy, we propose both adversarial and non-adversarial approaches to learn desirable matched joint distributions for unsupervised and supervised tasks. We unify a broad family of adversarial models as joint distribution matching problems. Our approach stabilizes learning of unsupervised bidirectional adversarial learning methods. Further, we introduce an extension for semi-supervised learning tasks. Theoretical results are validated in synthetic data and real-world applications.

Duke Authors

Cited Authors

  • Li, C; Liu, H; Chen, C; Pu, Y; Chen, L; Henao, R; Carin, L

Published Date

  • January 1, 2017

Published In

Volume / Issue

  • 2017-December /

Start / End Page

  • 5496 - 5504

International Standard Serial Number (ISSN)

  • 1049-5258

Citation Source

  • Scopus