Context-Specific Representation Abstraction for Deep Option Learning

TL;DR

This paper introduces CRADOL, a framework that enhances deep option learning by learning factored belief state representations, reducing policy search space, and improving sample efficiency in complex partially observable environments.

Contribution

CRADOL is a novel framework that combines temporal and context-specific representation abstraction to improve hierarchical reinforcement learning efficiency.

Findings

CRADOL significantly outperforms baselines in sample efficiency.

It effectively reduces the policy search space.

Demonstrates robustness in partially observable environments.

Abstract

Hierarchical reinforcement learning has focused on discovering temporally extended actions, such as options, that can provide benefits in problems requiring extensive exploration. One promising approach that learns these options end-to-end is the option-critic (OC) framework. We examine and show in this paper that OC does not decompose a problem into simpler sub-problems, but instead increases the size of the search over policy space with each option considering the entire state space during learning. This issue can result in practical limitations of this method, including sample inefficient learning. To address this problem, we introduce Context-Specific Representation Abstraction for Deep Option Learning (CRADOL), a new framework that considers both temporal abstraction and context-specific representation abstraction to effectively reduce the size of the search over policy space. Specifically, our method learns a factored belief state representation that enables each option to learn a policy over only a subsection of the state space. We test our method against hierarchical, non-hierarchical, and modular recurrent neural network baselines, demonstrating significant sample efficiency improvements in challenging partially observable environments.