A new method for discovering subgoals and constructing options in reinforcement learning

Please enable javascript in your browser.

Davoodabadi, M ; Sharif University of Technology

719 Viewed

Type of Document: Article
Abstract:
In this paper the problem of automatically discovering subtasks and hierarchies in reinforcement learning is considered. We present a novel method that allows an agent to autonomously discover subgoals and create a hierarchy from actions. Our method identifies subgoals by partitioning local state transition graphs. Options constructed for reaching these subgoals are added to action choices and used for accelerating the Q-Learning algorithm. Experimental results show significant performance improvements, especially in the initial learning phase
Keywords:
Autonomously discovering subgoals ; Community detection ; Option ; Hierarchical reinforcement learning ; Learning phase ; Local state ; Performance improvements ; Q-learning algorithms ; Subgoals ; Subtasks ; Artificial intelligence ; Learning algorithms ; Reinforcement learning
Source: Proceedings of the 5th Indian International Conference on Artificial Intelligence, IICAI 2011 ; 2011 , Pages 441-450 ; 9780972741286 (ISBN)