Authors:
Robert Wright
1
;
Steven Loscalzo
2
and
Lei Yu
3
Affiliations:
1
Air Force Research Lab, United States
;
2
Air Force Research Lab and Binghamton University, United States
;
3
Binghamton University, United States
Keyword(s):
Reinforcement learning, Feature selection, Neuroevolution.
Related
Ontology
Subjects/Areas/Topics:
Agents
;
Artificial Intelligence
;
Autonomous Systems
;
Bioinformatics
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computational Intelligence
;
Data Manipulation
;
Data Mining
;
Databases and Information Systems Integration
;
Enterprise Information Systems
;
Evolutionary Computing
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Information Systems Analysis and Specification
;
Knowledge Discovery and Information Retrieval
;
Knowledge-Based Systems
;
Machine Learning
;
Methodologies and Methods
;
Methodologies and Technologies
;
Neural Networks
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Operational Research
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Signal Processing
;
Simulation
;
Soft Computing
;
Symbolic Systems
;
Theory and Methods
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.