On-line Adaptative Curriculum Learning for GANs

Understand

Generative Adversarial Networks (GANs) can successfully approximate a probability distribution and produce realistic samples.

• However, open questions such as sufficient convergence conditions and mode collapse still persist.
• In this paper, we build on existing work in the area by proposing a novel framework for training the generator against an ensemble of discriminator networks, which can be seen as a one-student/multiple-teachers setting.
• We formalize this problem within the full-information adversarial bandit framework, where we evaluate the capability of an algorithm to select mixtures of discriminators for providing the generator with feedback during learning.