Unbiased Auxiliary Classifier GANs with MINE

Abstract

Auxiliary Classifier GANs (AC-GANs) are widely used conditional generative models and are capable of generating high-quality images. Previous work has pointed out that AC-GAN learns a biased distribution. To remedy this, Twin Auxiliary Classifier GAN (TAC-GAN) introduces a twin classifier to the min-max game. However, it has been reported that using a twin auxiliary classifier may cause instability in training. To this end, we propose an Unbiased Auxiliary GANs (UAC-GAN) that utilizes the Mutual Information Neural Estimator (MINE) to estimate the mutual information between the generated data distribution and labels. To further improve the performance, we also propose a novel projection-based statistics network architecture for MINE. Experimental results on three datasets, including Mixture of Gaussian (MoG), MNIST and CIFAR10 datasets, show that our UAC-GAN performs better than AC-GAN and TAC-GAN. Code can be found on the project website.

Accepted at CVPRW-20

References in corpus (6)

• Conditional Generative Adversarial Nets
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Improved Techniques for Training GANs
• cGANs with Projection Discriminator
• CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training

Cited by in corpus (5)

• Review: Deep Learning in Electron Microscopy
• Conditional GANs with Auxiliary Discriminative Classifier
• AE-StyleGAN: Improved Training of Style-Based Auto-Encoders
• Dual Projection Generative Adversarial Networks for Conditional Image Generation
• Label Geometry Aware Discriminator for Conditional Generative Networks