cGANs with Multi-Hinge Loss

2019/12/09 by Kavalerov, Ilya, Czaja, Wojciech, Chellappa, Rama
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1912.04216

Abstract

We propose a new algorithm to incorporate class conditional information into the critic of GANs via a multi-class generalization of the commonly used Hinge loss that is compatible with both supervised and semi-supervised settings. We study the compromise between training a state of the art generator and an accurate classifier simultaneously, and propose a way to use our algorithm to measure the degree to which a generator and critic are class conditional. We show the trade-off between a generator-critic pair respecting class conditioning inputs and generating the highest quality images. With our multi-hinge loss modification we are able to improve Inception Scores and Frechet Inception Distance on the Imagenet dataset. We make our tensorflow code available at https://github.com/ilyakava/gan.

Related

• cGANs with Projection Discriminator
• Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training
• On the Evaluation of Conditional GANs
• Are conditional GANs explicitly conditional?
• Self-labeled Conditional GANs