Improved Training of Wasserstein GANs, Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, Aaron C. Courville, 2017Advances in Neural Information Processing Systems 30, Vol. 30DOI: 10.48550/arXiv.1704.00028 - 这篇论文介绍了带有梯度惩罚的Wasserstein GAN (WGAN-GP),并广泛推广了用于稳定GAN训练的双时间尺度更新规则 (TTUR),包括对判别器和生成器学习率以及Adam优化器beta参数的具体建议。