Deakin University
Browse

File(s) under permanent embargo

Dual discriminator generative adversarial nets

conference contribution
posted on 2017-01-01, 00:00 authored by Tu Dinh Nguyen, Trung Minh Le, H Vu, Quoc-Dinh Phung
© 2017 Neural information processing systems foundation. All rights reserved. We propose in this paper a novel approach to tackle the problem of mode collapse encountered in generative adversarial network (GAN). Our idea is intuitive but proven to be very effective, especially in addressing some key limitations of GAN. In essence, it combines the Kullback-Leibler (KL) and reverse KL divergences into a unified objective function, thus it exploits the complementary statistical properties from these divergences to effectively diversify the estimated density in capturing multi-modes. We term our method dual discriminator generative adversarial nets (D2GAN) which, unlike GAN, has two discriminators; and together with a generator, it also has the analogy of a minimax game, wherein a discriminator rewards high scores for samples from data distribution whilst another discriminator, conversely, favoring data from the generator, and the generator produces data to fool both two discriminators. We develop theoretical analysis to show that, given the maximal discriminators, optimizing the generator of D2GAN reduces to minimizing both KL and reverse KL divergences between data distribution and the distribution induced from the data generated by the generator, hence effectively avoiding the mode collapsing problem. We conduct extensive experiments on synthetic and real-world large-scale datasets (MNIST, CIFAR-10, STL-10, ImageNet), where we have made our best effort to compare our D2GAN with the latest state-of-the-art GAN's variants in comprehensive qualitative and quantitative evaluations. The experimental results demonstrate the competitive and superior performance of our approach in generating good quality and diverse samples over baselines, and the capability of our method to scale up to ImageNet database.

History

Event

Neural Information Processing Systems. Conference (2017 : 31st : Long Beach, California)

Volume

30

Pagination

2671 - 2681

Publisher

Neural Information Processing Systems Foundation, Inc.

Location

Long Beach, California

Place of publication

[Long Beach, Calif.]

Start date

2017-12-04

End date

2017-12-09

ISSN

1049-5258

Language

eng

Publication classification

X Not reportable; EN Other conference paper

Title of proceedings

NIPS 2017 : Proceedings of the 31st Conference of Neural Information Processing Systems