EMBEDDED INCREMENTAL FEATURE SELECTION FOR REINFORCEMENT LEARNING
english
2011
Proceedings of the 3rd International Conference on Agents and Artificial Intelligence
unpublished
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to ABSTRACT Classical reinforcement learning techniques become impractical
more »
... in domains with large complex state spaces. The size of a domain's state space is dominated by the number of features used to describe the state. Fortunately, in many real-world environments learning an effective policy does not usually require all the provided features. In this paper we present a feature selection algorithm for reinforcement learning called Incremental Feature Selection Embedded in NEAT (IFSE-NEAT) that incorporates sequential forward search into neuroevolutionary algorithm NEAT. We provide an empirical analysis on a realistic simulated domain with many irrelevant and relevant features. Our results demonstrate that IFSE-NEAT selects smaller and more effective feature sets than alternative approaches, NEAT and FS-NEAT, and superior performance characteristics as the number of available features increases. 15. SUBJECT TERMS Reinforcement Learning, Feature Selection, Neuroevolution 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT UU 18. NUMBER OF PAGES 7 19a. NAME OF RESPONSIBLE PERSON ALEX F. SISTI a. REPORT U b. ABSTRACT U c. THIS PAGE U 19b. TELEPHONE NUMBER (Include area code) N/A Standard Form 298 (Rev. 8-98) Abstract: Classical reinforcement learning techniques become impractical in domains with large complex state spaces. The size of a domain's state space is dominated by the number of features used to describe the state. Fortunately, in many real-world environments learning an effective policy does not usually require all the provided features. In this paper we present a feature selection algorithm for reinforcement learning called Incremental Feature Selection Embedded in NEAT (IFSE-NEAT) that incorporates sequential forward search into neuroevolutionary algorithm NEAT. We provide an empirical analysis on a realistic simulated domain with many irrelevant and relevant features. Our results demonstrate that IFSE-NEAT selects smaller and more effective feature sets than alternative approaches, NEAT and FS-NEAT, and superior performance characteristics as the number of available features increases.
doi:10.5220/0003153402630268
fatcat:obfu5xy6i5dnzjfqkisnfe2saa