papers AI Learner
The Github is limit! Click to go to the new site.

Dyna-AIL : Adversarial Imitation Learning by Planning

2019-03-08
Vaibhav Saxena, Srinivasan Sivanandan, Pulkit Mathur

Abstract

Adversarial methods for imitation learning have been shown to perform well on various control tasks. However, they require a large number of environment interactions for convergence. In this paper, we propose an end-to-end differentiable adversarial imitation learning algorithm in a Dyna-like framework for switching between model-based planning and model-free learning from expert data. Our results on both discrete and continuous environments show that our approach of using model-based planning along with model-free learning converges to an optimal policy with fewer number of environment interactions in comparison to the state-of-the-art learning methods.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1903.03234

PDF

http://arxiv.org/pdf/1903.03234


Similar Posts

Comments