Learning Robust Options

Authors

  • Daniel Mankowitz Technion Israel Institute of Technology
  • Timothy Mann Google Deepmind
  • Pierre-Luc Bacon McGill University
  • Doina Precup McGill University
  • Shie Mannor Technion Israel Institute of Technology

Keywords:

Reinforcement Learning, Robustness

Abstract

Robust reinforcement learning aims to produce policies that have strong guarantees even in the face of environments/transition models whose parameters have strong uncertainty. Existing work uses value-based methods and the usual primitive action setting. In this paper, we propose robust methods for learning temporally abstract actions, in the framework of options. We present a Robust Options Policy Iteration (ROPI) algorithm with convergence guarantees, which learns options that are robust to model uncertainty. We utilize ROPI to learn robust options with the Robust Options Deep Q Network (RO-DQN) that solves multiple tasks and mitigates model misspecification due to model uncertainty. We present experimental results which suggest that policy iteration with linear features may have an inherent form of robustness when using coarse feature representations. In addition, we present experimental results which demonstrate that robustness helps policy iteration implemented on top of deep neural networks to generalize over a much broader range of dynamics than non-robust policy iteration.

Downloads

Published

2018-04-26

How to Cite

Mankowitz, D., Mann, T., Bacon, P.-L., Precup, D., & Mannor, S. (2018). Learning Robust Options. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1). Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/12115

Issue

Section

AAAI Technical Track: Reasoning under Uncertainty