Consistency Regularization for Generative Adversarial Networks

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.

ICLR2020

References in corpus (12)

• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Improved Regularization of Convolutional Neural Networks with Cutout
• Self-Attention Generative Adversarial Networks
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Unsupervised Data Augmentation for Consistency Training
• Temporal Ensembling for Semi-Supervised Learning
• MixMatch: A Holistic Approach to Semi-Supervised Learning
• Learning with Pseudo-Ensembles
• On Convergence and Stability of GANs
• Hierarchical Implicit Models and Likelihood-Free Variational Inference
• Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network

Cited by in corpus (11)

• TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up
• An Overview of Deep Semi-Supervised Learning
• A survey of synthetic data augmentation methods in computer vision
• A Systematic Survey of Regularization and Normalization in GANs
• PConv: Simple yet Effective Convolutional Layer for Generative Adversarial Network
• Understanding and Stabilizing GANs' Training Dynamics with Control Theory
• Attentive Normalization for Conditional Image Generation
• Bridge the Domain Gap Between Ultra-wide-field and Traditional Fundus Images via Adversarial Domain Adaptation
• Transfer Learning with Pre-trained Conditional Generative Models
• On the Guaranteed Almost Equivalence between Imitation Learning from Observation and Demonstration
• Class Balancing GAN with a Classifier in the Loop