Training Generative Adversarial Networks Via Turing Test
This addresses training efficiency and scalability issues for GANs in generative modeling, though it appears incremental as it builds on existing GAN frameworks.
The paper tackles the problem of training Generative Adversarial Networks (GANs) by introducing a new adversarial pattern based on a Turing test interpretation, which accelerates training and enables success on 256x256 resolution without careful hyperparameter tuning.
In this article, we introduce a new mode for training Generative Adversarial Networks (GANs). Rather than minimizing the distance of evidence distribution $\tilde{p}(x)$ and the generative distribution $q(x)$, we minimize the distance of $\tilde{p}(x_r)q(x_f)$ and $\tilde{p}(x_f)q(x_r)$. This adversarial pattern can be interpreted as a Turing test in GANs. It allows us to use information of real samples during training generator and accelerates the whole training procedure. We even find that just proportionally increasing the size of discriminator and generator, it succeeds on 256x256 resolution without adjusting hyperparameters carefully.