MultiselfGAN: A Self-Guiding Neural Architecture Search Method for Generative Adversarial Networks With Multicontrollers

In recent years, reinforcement learning (RL) and gradient optimization were applied with neural architecture search algorithms in the generative adversarial network (GAN) to achieve their state-of-the-art (SOTA) performance. However, the existing RL-based methods utilized the calculation of inception score or Fréchet inception distance as the reward value to guide the controller, which actually wasted much of searching time. In order to improve the search efficiency without degradation of performance, this article proposes recycling the discriminator to evaluate the performance of architectures, in other words, we propose to self-guide the search process. In the mean time, we introduce new concept of multiple controllers and the method of reward shaping to independently and effectively search the cell architectures. The experiments demonstrate the effectiveness and efficiency of our multiself GAN and the ablation study also exhibits its robustness.