As social media grows faster, harassment becomes more prevalent which leads
to considered fake detection a fascinating field among researchers. The graph
nature of data with the large number of nodes caused different obstacles
including a considerable amount of unrelated features in matrices as high
dispersion and imbalance classes in the dataset. To deal with these issues
Auto-encoders and a combination of semi-supervised learning and the GAN
algorithm which is called SGAN were used. This paper is deploying a smaller
number of labels and applying SGAN as a classifier. The result of this test
showed that the accuracy had reached 91\% in detecting fake accounts using only
100 labeled samples.

本研究使用自动编码器和半监督学习与 GAN 算法的组合（SGAN），并将其作为分类器，仅使用 100 个标记样本即可将精度提高达 91％，以检测虚假账户，适用于社交媒体的大规模数据。

半监督学习与 GAN 在不平衡数据集中的伪造检测

Fake detection in imbalance dataset by Semi-supervised learning with GAN

We know SGAN may have a risk of gradient vanishing. A significant improvement
is WGAN, with the help of 1-Lipschitz constraint on discriminator to prevent
from gradient vanishing. Is there any GAN having no gradient vanishing and no
1-Lipschitz constraint on discriminator? We do find one, called GAN-QP.
To construct a new framework of Generative Adversarial Network (GAN) usually
includes three steps: 1. choose a probability divergence; 2. convert it into a
dual form; 3. play a min-max game. In this articles, we demonstrate that the
first step is not necessary. We can analyse the property of divergence and even
construct new divergence in dual space directly. As a reward, we obtain a
simpler alternative of WGAN: GAN-QP. We demonstrate that GAN-QP have a better
performance than WGAN in theory and practice.

本文提出了一种新型的生成对抗网络 GAN-QP，它不需要 Lipschitz 约束，也不需要选择概率分布，可以直接在对偶空间中构建新的分布，并且理论和实践中要优于 WGAN。

GAN-QP: 一种无梯度消失和 Lipschitz 约束的新型 GAN 框架

GAN-QP: A Novel GAN Framework without Gradient Vanishing and Lipschitz  Constraint

The Generative Adversarial Networks (GANs) have demonstrated impressive
performance for data synthesis, and are now used in a wide range of computer
vision tasks. In spite of this success, they gained a reputation for being
difficult to train, what results in a time-consuming and human-involved
development process to use them.
We consider an alternative training process, named SGAN, in which several
adversarial "local" pairs of networks are trained independently so that a
"global" supervising pair of networks can be trained against them. The goal is
to train the global pair with the corresponding ensemble opponent for improved
performances in terms of mode coverage. This approach aims at increasing the
chances that learning will not stop for the global pair, preventing both to be
trapped in an unsatisfactory local minimum, or to face oscillations often
observed in practice. To guarantee the latter, the global pair never affects
the local ones.
The rules of SGAN training are thus as follows: the global generator and
discriminator are trained using the local discriminators and generators,
respectively, whereas the local networks are trained with their fixed local
opponent.
Experimental results on both toy and real-world problems demonstrate that
this approach outperforms standard training in terms of better mitigating mode
collapse, stability while converging and that it surprisingly, increases the
convergence speed as well.

本研究探讨了一种名为 SGAN 的备选训练方法，通过该方法，多个对抗的 “本地” 网络对可独立进行训练，以便 “全局” 监督网络可以针对它们进行训练，以提高模式覆盖性能。实验结果表明，该方法在减轻模式崩溃、稳定收敛和加速收敛方面都表现出优于标准训练的性能。