Consistency Regularization for Generative Adversarial Networks

2019/10/26 by Han Zhang, Zhang, Han, Zizhao Zhang +5 · 4 citations
Computer Science · #Advanced Image Processing Techniques #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.1910.12027

openalex publication_date 2019/10/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Generative Adversarial Networks (GANs) are known to be difficult to train, despite considerable research effort. Several regularization techniques for stabilizing training have been proposed, but they introduce non-trivial computational overheads and interact poorly with existing techniques like spectral normalization. In this work, we propose a simple, effective training stabilizer based on the notion of consistency regularization---a popular technique in the semi-supervised learning literature. In particular, we augment data passing into the GAN discriminator and penalize the sensitivity of the discriminator to these augmentations. We conduct a series of experiments to demonstrate that consistency regularization works effectively with spectral normalization and various GAN architectures, loss functions and optimizer settings. Our method achieves the best FID scores for unconditional image generation compared to other regularization methods on CIFAR-10 and CelebA. Moreover, Our consistency regularized GAN (CR-GAN) improves state-of-the-art FID scores for conditional generation from 14.73 to 11.48 on CIFAR-10 and from 8.73 to 6.66 on ImageNet-2012.

Citations

• ImageNet Large Scale Visual Recognition Challenge
• Conditional Image Synthesis With Auxiliary Classifier GANs
• On Convergence and Stability of GANs
• Self-Attention Generative Adversarial Networks
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• MixMatch: A Holistic Approach to Semi-Supervised Learning
• Temporal Ensembling for Semi-Supervised Learning
• Unsupervised Data Augmentation for Consistency Training
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Learning Multiple Layers of Features from Tiny Images

Cited by

• Learning from the Undesirable: Robust Adaptation of Language Models without Forgetting
• SONA: Learning Conditional, Unconditional, and Mismatching-Aware Discriminator
• Optimal Multi-Task Learning at Regularization Horizon for Speech Translation Task
• TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up

Related

• Improved Consistency Regularization for GANs
• Stabilizing Training of Generative Adversarial Networks through Regularization
• Consistency Regularization with Generative Adversarial Networks for Semi-Supervised Learning
• Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studies
• Regularizing Generative Adversarial Networks under Limited Data