Progressive Augmentation of GANs

NeurIPS 2019  ·  Dan Zhang, Anna Khoreva ·

Training of Generative Adversarial Networks (GANs) is notoriously fragile, requiring to maintain a careful balance between the generator and the discriminator in order to perform well. To mitigate this issue we introduce a new regularization technique - progressive augmentation of GANs (PA-GAN). The key idea is to gradually increase the task difficulty of the discriminator by progressively augmenting its input or feature space, thus enabling continuous learning of the generator. We show that the proposed progressive augmentation preserves the original GAN objective, does not compromise the discriminator's optimality and encourages a healthy competition between the generator and discriminator, leading to the better-performing generator. We experimentally demonstrate the effectiveness of PA-GAN across different architectures and on multiple benchmarks for the image synthesis task, on average achieving ~3 point improvement of the FID score.

PDF Abstract NeurIPS 2019 PDF NeurIPS 2019 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Generation CelebA-HQ 128x128 PA-GAN FID 15.4 # 3
Image Generation CIFAR-10 PA-GAN FID 16.1 # 112

Methods