Please use this identifier to cite or link to this item: http://hdl.handle.net/10525/851

 Title: Local Goals Driven Hierarchical Reinforcement Learning Authors: Pchelkin, Arthur Keywords: Reinforcement LearningHierarchical BehaviourEfficient ExplorationPOMDPsNon-MarkovLocal GoalsInternal RewardSubgoal Learning Issue Date: 2004 Publisher: Institute of Information Theories and Applications FOI ITHEA Abstract: Efficient exploration is of fundamental importance for autonomous agents that learn to act. Previous approaches to exploration in reinforcement learning usually address exploration in the case when the environment is fully observable. In contrast, the current paper, like the previous paper [Pch2003], studies the case when the environment is only partially observable. One additional difficulty is considered – complex temporal dependencies. In order to overcome this additional difficulty a new hierarchical reinforcement learning algorithm is proposed. The learning algorithm exploits a very simple learning principle, similar to Q-learning, except the lookup table has one more variable – the currently selected goal. Additionally, the algorithm uses the idea of internal reward for achieving hard-to-reach states [Pch2003]. The proposed learning algorithm is experimentally investigated in partially observable maze problems where it shows a robust ability to learn a good policy. Description: * This research was partially supported by the Latvian Science Foundation under grant No.02-86d. URI: http://hdl.handle.net/10525/851 ISSN: 1313-0463 Appears in Collections: Volume 11 Number 1

Files in This Item:

File Description SizeFormat