Coevolution of Generative Adversarial Networks
This work addresses training challenges in GANs for computer vision applications, but it is incremental as it builds on existing neuroevolution and coevolution techniques.
The paper tackles training stability and architecture design issues in GANs by proposing COEGAN, which combines neuroevolution and coevolution, resulting in improved stability and automatic discovery of efficient architectures on the MNIST dataset, partially solving mode collapse.
Generative adversarial networks (GAN) became a hot topic, presenting impressive results in the field of computer vision. However, there are still open problems with the GAN model, such as the training stability and the hand-design of architectures. Neuroevolution is a technique that can be used to provide the automatic design of network architectures even in large search spaces as in deep neural networks. Therefore, this project proposes COEGAN, a model that combines neuroevolution and coevolution in the coordination of the GAN training algorithm. The proposal uses the adversarial characteristic between the generator and discriminator components to design an algorithm using coevolution techniques. Our proposal was evaluated in the MNIST dataset. The results suggest the improvement of the training stability and the automatic discovery of efficient network architectures for GANs. Our model also partially solves the mode collapse problem.