Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect
Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect
复制标题
DOI：
--
复制
发表时间：
2018-02
期刊：
影响因子：
--
通讯作者：
Xiang Wei;Boqing Gong;Zixia Liu;W. Lu;Liqiang Wang
中科院分区：
文献类型：
--
作者：
Xiang Wei;Boqing Gong;Zixia Liu;W. Lu;Liqiang Wang
尽管对各种问题和应用都有影响，但生成对抗网络（GAN）非常难以训练。\cite{arjovsky 2017 towards}正式分析了这个问题，他还提出了一个替代方向，以避免GAN最小最大双人训练中的警告。相应的算法，称为Wasserstein GAN（WGAN），取决于1-Lipschitz连续性。在本文中，我们提出了一种新的方法来加强WGAN训练过程中的Lipschitz连续性。我们的方法将WGAN与最近的一种半监督学习方法无缝连接起来。因此，它不仅产生了比以前的方法更好的照片般逼真的样本，而且还产生了最先进的半监督学习结果。特别是，我们的方法仅使用1，000张CIFAR-10图像就获得了超过5.0的初始分数，并且是第一个仅使用4，000张标记图像就超过CIFAR-10数据集90%准确度的方法。
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.