Simplified Frechet Distance for Generative Adversarial Nets
Simplified Frechet Distance for Generative Adversarial Nets
复制标题
DOI：
10.3390/s20061548
复制
发表时间：
2020-03-01
期刊：
影响因子：
3.9
通讯作者：
Hwang, Eenjun
中科院分区：
文献类型：
--
作者：
Kim, Chung-Il;Kim, Meejoung;Hwang, Eenjun
我们引入了两个分布之间的距离度量，并提出了一种生成对抗网络（GAN）模型：简化Frechet距离（SFD）和简化Frechet GAN （SFGAN）。虽然通过GAN生成的数据与真实数据相似，但由于GAN的对抗性结构，其训练往往不稳定。一个可能的解决方案是考虑Frechet距离（FD）。然而，由于FD存在协方差项，因此无法实现。SFD克服了复杂性，使我们能够在网络中实现。SFGAN的结构基于边界平衡GAN (BEGAN)，并在损失函数中使用SFD。在CelebA和CIFAR-10等数据集上进行了实验。将SFGAN和begin的损失和生成的样本与几个距离度量进行了比较。对于SFGAN来说，模式崩溃和/或模式下降的证据直到3000k步才发生，而对于BEGAN来说，模式崩溃和/或模式下降发生在457k到968k步之间。实验结果表明，与其他距离度量相比，SFD使gan更加稳定，并且SFD弥补了基于began的网络结构模型的不足。根据实验结果，我们可以得出结论，SFD比其他指标更适合GAN。
We introduce a distance metric between two distributions and propose a Generative Adversarial Network (GAN) model: the Simplified Frechet distance (SFD) and the Simplified Frechet GAN (SFGAN). Although the data generated through GANs are similar to real data, GAN often undergoes unstable training due to its adversarial structure. A possible solution to this problem is considering Frechet distance (FD). However, FD is unfeasible to realize due to its covariance term. SFD overcomes the complexity so that it enables us to realize in networks. The structure of SFGAN is based on the Boundary Equilibrium GAN (BEGAN) while using SFD in loss functions. Experiments are conducted with several datasets, including CelebA and CIFAR-10. The losses and generated samples of SFGAN and BEGAN are compared with several distance metrics. The evidence of mode collapse and/or mode drop does not occur until 3000k steps for SFGAN, while it occurs between 457k and 968k steps for BEGAN. Experimental results show that SFD makes GANs more stable than other distance metrics used in GANs, and SFD compensates for the weakness of models based on BEGAN-based network structure. Based on the experimental results, we can conclude that SFD is more suitable for GAN than other metrics.