Flexible Option Learning

Martin Klissarov, Doina Precup

Keywords: temporal abstraction, options, hierarchical reinforcement learning, deep reinforcement learning, reinforcement learning

TL;DR: We propose updates that allow for learning all relevant options simulteneously without introducing additional estimators. We verify that our approach can improve sample efficiency and can open the door to more flexibility when learning options.

Abstract: Temporal abstraction in reinforcement learning (RL), offers the promise of improving generalization and knowledge transfer in complex environments, by propagating information more efficiently over time. Although option learning was initially formulated in a way that allows updating many options simultaneously, using off-policy, intra-option learning (Sutton, Precup & Singh, 1999) , many of the recent hierarchical reinforcement learning approaches only update a single option at a time: the option currently executing. We revisit and extend intra-option learning in the context of deep reinforcement learning, in order to enable updating all options consistent with current primitive action choices, without introducing any additional estimates. Our method can therefore be naturally adopted in most hierarchical RL frameworks. When we combine our approach with the option-critic algorithm for option discovery, we obtain significant improvements in performance and data-efficiency across a wide variety of domains.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/mklissa/MOC

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2112.03097/code)