Generative Adversarial Network for Imitation Learning from Single Demonstration

Authors

Tho Nguyen Duc, School of Engineering and Science, Shibaura Institute of Technology, Japan.Follow
Chanh Minh Tran, School of Engineering and Science, Shibaura Institute of Technology, JapanFollow
Phan Xuan Tan, School of Engineering and Science, Shibaura Institute of Technology, Japan.Follow
Eiji Kamioka, School of Engineering and Science, Shibaura Institute of Technology, JapanFollow

Abstract

Imitation learning is an effective method for training an autonomous agent to accomplish a task by imitating expert behaviors in their demonstrations. However, traditional imitation learning methods require a large number of expert demonstrations in order to learn a complex behavior. Such a disadvantage has limited the potential of imitation learning in complex tasks where the expert demonstrations are not sufficient. In order to address the problem, we propose a Generative Adversarial Network-based model which is designed to learn optimal policies using only a single demonstration. The proposed model is evaluated on two simulated tasks in comparison with other methods. The results show that our proposed model is capable of completing considered tasks despite the limitation in the number of expert demonstrations, which clearly indicate the potential of our model.

Keywords

Deep Learning, Few-shot Learning, Generative Adversarial Network, Imitation Learning, One-shot Learning

Article Type

Supplemental Issue

How to Cite this Article

Duc, Tho Nguyen; Tran, Chanh Minh; Tan, Phan Xuan; and Kamioka, Eiji (2021) "Generative Adversarial Network for Imitation Learning from Single Demonstration," Baghdad Science Journal: Vol. 18: Iss. 4, Article 46.
DOI: https://doi.org/10.21123/bsj.2021.18.4(Suppl.).1350