Using Novelty Seeking Reward Evolution Strategies to Train Generative Adversarial Networks

Jabr, Khaled

Using Novelty Seeking Reward Evolution Strategies to Train Generative Adversarial Networks

dc.contributor.advisor	McGovern, Amy
dc.contributor.author	Jabr, Khaled
dc.contributor.committeeMember	Hougen, Dean
dc.contributor.committeeMember	Fagg, Andrew
dc.date.accessioned	2018-12-14T22:22:38Z
dc.date.available	2018-12-14T22:22:38Z
dc.date.issued	2018-12-14
dc.date.manuscript	2018-12-14
dc.description.abstract	Generative Adversarial Networks (GANs) are a subclass of deep generative models that aim to implicitly learn to model a data distribution. While GANs have gained wide research attention, and achieved much success, when trained with first-order stochastic gradient descent (SGD), they suffer from training instabilities, such as non-convergence and mode collapse, in which they fail to converge to the Nash equilibrium of the minimax game, and fail to learn all the modes of the data distribution, where the samples of the generator lack diversity. To this end, this thesis investigates the use of evolution strategies (ES) to train GANs, and address the mode collapse issue. The evolution strategies (ES) algorithm used in this work is simplified version of natural evolution strategies (NES). ES achieved very impressive and competitive results against state of the art SGD-based deep reinforcement learning (RL) algorithms. A quality diversity hybrid of ES, known as Novelty Seeking Reward Evolution Strategies (NSR-ES), that aims to encourage exploration and diversity is particularly interesting in relation to the mode collapse problem is also used. In this work we propose two algorithms to train GANs, ES-GAN and NSR-ES-GAN, and we carryout experimentation on a constrained GAN setup where mode collapse is well known to study how our algorithms can help overcome the issue. Our results show that using ES and NSR-ES to train GANs fails to overcome the mode collapse issue, and suggests that more robust and domain specific techniques are needed to overcome the problem.	en_US
dc.identifier.uri	https://hdl.handle.net/11244/316799
dc.language	en	en_US
dc.subject	Generative Adversarial Networks	en_US
dc.subject	Evolution Strategies	en_US
dc.subject	Novelty Search	en_US
dc.subject	Neuroevolution	en_US
dc.thesis.degree	Master of Science	en_US
dc.title	Using Novelty Seeking Reward Evolution Strategies to Train Generative Adversarial Networks	en_US
ou.group	Gallogly College of Engineering::School of Computer Science	en_US