Learning Representations in Model-Free Hierarchical Reinforcement LearningDownload PDF

27 Sept 2018 (modified: 05 May 2023)ICLR 2019 Conference Blind SubmissionReaders: Everyone
Abstract: Common approaches to Reinforcement Learning (RL) are seriously challenged by large-scale applications involving huge state spaces and sparse delayed reward feedback. Hierarchical Reinforcement Learning (HRL) methods attempt to address this scalability issue by learning action selection policies at multiple levels of temporal abstraction. Abstraction can be had by identifying a relatively small set of states that are likely to be useful as subgoals, in concert with the learning of corresponding skill policies to achieve those subgoals. Many approaches to subgoal discovery in HRL depend on the analysis of a model of the environment, but the need to learn such a model introduces its own problems of scale. Once subgoals are identified, skills may be learned through intrinsic motivation, introducing an internal reward signal marking subgoal attainment. In this paper, we present a novel model-free method for subgoal discovery using incremental unsupervised learning over a small memory of the most recent experiences of the agent. When combined with an intrinsic motivation learning mechanism, this method learns subgoals and skills together, based on experiences in the environment. Thus, we offer an original approach to HRL that does not require the acquisition of a model of the environment, suitable for large-scale applications. We demonstrate the efficiency of our method on two RL problems with sparse delayed feedback: a variant of the rooms environment and the ATARI 2600 game called Montezuma's Revenge.
Keywords: Reinforcement Learning, Model-Free Hierarchical Reinforcement Learning, Subgoal Discovery, Unsupervised Learning, Temporal Difference, Temporal Abstraction, Intrinsic Motivation, Markov Decision Processes, Deep Reinforcement Learning, Optimization
TL;DR: We offer an original approach to model-free deep hierarchical reinforcement learning, including unsupervised subgoal discovery and unified temporal abstraction and intrinsic motivation learning.
4 Replies

Loading