Twin Auxiliary Classifiers GAN

Mingming Gong, Yanwu Xu, Chuyuan Li, Kun Zhang, Kayhan Batmanghelich.  Twin Auxiliary Classifiers GAN. 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada

Conditional generative models enjoy remarkable progress over the past few years. One of the popular conditional models is Auxiliary Classifier GAN (AC-GAN), which generates highly discriminative images by extending the loss function of GAN with an auxiliary classifier. However, the diversity of the generated samples by AC-GAN tends to decrease as the number of classes increases, hence limiting its power on large-scale data. In this paper, we identify the source of the low diversity issue theoretically and propose a practical solution to solve the problem. We show that the auxiliary classifier in AC-GAN imposes perfect separability, which is disadvantageous when the supports of the class distributions have significant overlap. To address the issue, we propose Twin Auxiliary Classifiers Generative Adversarial Net (TAC-GAN) that further benefits from a new player that interacts with other players (the generator and the discriminator) in GAN. Theoretically, we demonstrate that TAC-GAN can effectively minimize the divergence between the generated and real-data distributions. Extensive experimental results show that our TAC-GAN can successfully replicate the true data distributions on simulated data, and significantly improves the diversity of class-conditional image generation on real datasets.

Publication Year: 
2019
Faculty Author: 
Publication Credits: 
Mingming Gong, Yanwu Xu, Chuyuan Li, Kun Zhang, Kayhan Batmanghelich. 
Publication Download: 
^