Generative Adversarial Networks (GANs) are increasingly adopted by the industry to synthesize realistic images. Due to data not being centrally available, Multi-Discriminator (MD)-GANs training framework employs multiple discriminators that have direct access to the real data. Distributedly training a joint GAN model entails the risk of free-riders, i.e., participants that aim to benefit from the common model while only pretending to participate in the training process. In this paper, we conduct the first characterization study of the impact of free-riders on MD-GAN. Based on two production prototypes of MD-GAN, we find that free-riders drastically reduce the ability of MD-GANs to produce images that are indistinguishable from real data, i.e., they increase the FID score -- the standard measure to assess the quality of generated images. To mitigate the model degradation, we propose a defense strategy against free-riders in MD-GAN, termed DFG. DFG distinguishes free-riders and benign participants through periodic probing and clustering of discriminators' responses based on a reference response of free-riders, which then allows the generator to exclude the detected free-riders from the training. Furthermore, we extend our defense, termed DFG+, to enable discriminators to filter out free-riders at the variant of MD-GAN that allows peer exchanges of discriminators networks. Extensive evaluation on various scenarios of free-riders, MD-GAN architecture, and three datasets show that our defenses effectively detect free-riders. With 1 to 5 free-riders, DFG and DFG+ averagely decreases FID by 5.22% to 11.53% for CIFAR10 and 5.79% to 13.22% for CIFAR100 in comparison to an attack without defense. In a shell, the proposed DFG(+) can effectively defend against free-riders without affecting benign clients at a negligible computation overhead.
翻译:行业越来越多地采用GAN(GAN)来合成现实图像。由于数据无法集中提供,多分解器-GAN(MD)培训框架使用多种直接获取真实数据的歧视者。分布式培训GAN(GAN)模式带来免费搭车者的风险,即,参与者的目的是从共同模式中受益,而只是假装参与培训过程。在本文中,我们首次对免费搭车者对MD-GAN的影响进行了定性研究。基于MD-GAN(MD-GAN)的两种生产模型,我们有效地发现免费搭车者极大地降低了MD-GAN(MD-GAAN)制作与真实数据直接接触的图像的能力。即,它们增加了FID的评分 -- -- 评估所生成图像质量的标准衡量标准。为了减轻模型的退化,我们建议对MD-G(MD-G)的免费搭车者进行防御战略攻击,DFG(DFG)将自由驾驶者和良性参与者在MDG(M-G)网络中进行区分,我们通过定期试算和直接测试数据组合,让免费的服务器(DRC)进行自由评估,让我们免费的DNA(G)的R-G(O-G)得到免费的响应)。