Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect

Abstract

Despite being impactful on a variety of problems and applications, the generative adversarial nets (GANs) are remarkably difficult to train. This issue is formally analyzed by \cite{arjovsky2017towards}, who also propose an alternative direction to avoid the caveats in the minmax two-player training of GANs. The corresponding algorithm, called Wasserstein GAN (WGAN), hinges on the 1-Lipschitz continuity of the discriminator. In this paper, we propose a novel approach to enforcing the Lipschitz continuity in the training procedure of WGANs. Our approach seamlessly connects WGAN with one of the recent semi-supervised learning methods. As a result, it gives rise to not only better photo-realistic samples than the previous methods but also state-of-the-art semi-supervised learning results. In particular, our approach gives rise to the inception score of more than 5.0 with only 1,000 CIFAR-10 images and is the first that exceeds the accuracy of 90% on the CIFAR-10 dataset using only 4,000 labeled images, to the best of our knowledge.

arXiv commentAccepted as a conference paper in International Conference on Learning Representation(ICLR). Xiang Wei and Boqing Gong contributed equally in this work

References in corpus (23)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Temporal Ensembling for Semi-Supervised Learning
• Improved Training of Wasserstein GANs
• Improved Techniques for Training GANs
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• BEGAN: Boundary Equilibrium Generative Adversarial Networks
• Adversarially Learned Inference
• MMD GAN: Towards Deeper Understanding of Moment Matching Network
• Generative Moment Matching Networks
• Towards Principled Methods for Training Generative Adversarial Networks
• Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks
• Triple Generative Adversarial Nets
• Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
• Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities
• Semi-supervised Learning with GANs: Manifold Invariance with Improved Inference
• Learning to Draw Samples: With Application to Amortized MLE for Generative Adversarial Learning
• Stacked Generative Adversarial Networks
• Towards Principled Unsupervised Learning
• f-GANs in an Information Geometric Nutshell

Cited by in corpus (19)

• A Systematic Survey of Regularization and Normalization in GANs
• Deepfake Detection Scheme Based on Vision Transformer and Distillation
• Optimal transport mapping via input convex neural networks
• Improved Consistency Regularization for GANs
• Cleaning our own Dust: Simulating and Separating Galactic Dust Foregrounds with Neural Networks
• 2-Wasserstein Approximation via Restricted Convex Potentials with Application to Improved Training for GANs
• An error analysis of generative adversarial networks for learning distributions
• O-GAN: Extremely Concise Approach for Auto-Encoding Generative Adversarial Networks
• Generative Convolution Layer for Image Generation
• Improving GAN Training with Probability Ratio Clipping and Sample Reweighting
• Accelerating Monte Carlo Bayesian Inference via Approximating Predictive Uncertainty over Simplex
• Forecasting Spatio-Temporal Renewable Scenarios: a Deep Generative Approach
• Out-of-Sample Extrapolation with Neuron Editing
• Approximating Lipschitz continuous functions with GroupSort neural networks
• Privacy Enhancing Machine Learning via Removal of Unwanted Dependencies
• Artist Style Transfer Via Quadratic Potential
• Consistency Regularization with Generative Adversarial Networks for Semi-Supervised Learning
• Sliced Wasserstein Generative Models
• Improving Object Detection with Selective Self-supervised Self-training