Discriminator optimal transport
Discriminator optimal transport
复制标题
DOI：
--
复制
发表时间：
2019-10
期刊：
影响因子：
--
通讯作者：
A. Tanaka
中科院分区：
文献类型：
--
作者：
A. Tanaka
在一类广泛的生成对抗网络中，我们证明了鉴别器优化过程增加了目标分布$p$和生成器分布$p_G$之间的Wasserstein距离的对偶代价函数的下界。这意味着训练的鉴别器可以近似从$p_G$到$p$的最优传输（OT）。基于一些实验和一些OT理论，我们提出了鉴别器最优传输（DOT）方案来改善生成的图像。我们证明，它提高了由CIFAR-10、STL-10和ImageNet训练的条件GAN的公共预训练模型计算的初始分数和FID。
Within a broad class of generative adversarial networks, we show that discriminator optimization process increases a lower bound of the dual cost function for the Wasserstein distance between the target distribution $p$ and the generator distribution $p_G$. It implies that the trained discriminator can approximate optimal transport (OT) from $p_G$ to $p$. Based on some experiments and a bit of OT theory, we propose discriminator optimal transport (DOT) scheme to improve generated images. We show that it improves inception score and FID calculated by un-conditional GAN trained by CIFAR-10, STL-10 and a public pre-trained model of conditional GAN trained by ImageNet.