caa a22 4500 605514844 CHVBK 20210128100706.0 cr unu---uuuuu 210128e20150901xx s 000 0 eng 10.1007/s10458-014-9268-y doi (NATIONALLICENCE)springer-10.1007/s10458-014-9268-y Predictive feature selection for genetic policy search [Elektronische Daten] [Steven Loscalzo, Robert Wright, Lei Yu] Automatic learning of control policies is becoming increasingly important to allow autonomous agents to operate alongside, or in place of, humans in dangerous and fast-paced situations. Reinforcement learning (RL), including genetic policy search algorithms, comprise a promising technology area capable of learning such control policies. Unfortunately, RL techniques can take prohibitively long to learn a sufficiently good control policy in environments described by many sensors (features). We argue that in many cases only a subset of available features are needed to learn the task at hand, since others may represent irrelevant or redundant information. In this work, we propose a predictive feature selection framework that analyzes data obtained during execution of a genetic policy search algorithm to identify relevant features on-line. This serves to constrain the policy search space and reduces the time needed to locate a sufficiently good policy by embedding feature selection into the process of learning a control policy. We explore this framework through an instantiation called predictive feature selection embedded in neuroevolution of augmenting topology (NEAT), or PFS-NEAT. In an empirical study, we demonstrate that PFS-NEATis capable of enabling NEAT to successfully find good control policies in two benchmark environments, and show that it can outperform three competing feature selection algorithms, FS-NEAT, FD-NEAT, and SAFS-NEAT, in several variants of these environments. The Author(s), 2014 Genetic policy search nationallicence Feature selection nationallicence Dimensionality reduction nationallicence Reinforcement learning nationallicence Loscalzo Steven AFRL Information Directorate, 26 Electronic Parkway, 13441, Rome, NY, USA aut Wright Robert Binghamton University, 4400 Vestal Parkway East, 13902, Binghamton, NY, USA aut Yu Lei Binghamton University, 4400 Vestal Parkway East, 13902, Binghamton, NY, USA aut Autonomous Agents and Multi-Agent Systems Springer US; http://www.springer-ny.com 29/5(2015-09-01), 754-786 1387-2532 29:5<754 2015 29 10458 https://doi.org/10.1007/s10458-014-9268-y text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s10458-014-9268-y text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Loscalzo Steven AFRL Information Directorate, 26 Electronic Parkway, 13441, Rome, NY, USA aut NATIONALLICENCE 700 1- Wright Robert Binghamton University, 4400 Vestal Parkway East, 13902, Binghamton, NY, USA aut NATIONALLICENCE 700 1- Yu Lei Binghamton University, 4400 Vestal Parkway East, 13902, Binghamton, NY, USA aut NATIONALLICENCE 773 0- Autonomous Agents and Multi-Agent Systems Springer US; http://www.springer-ny.com 29/5(2015-09-01), 754-786 1387-2532 29:5<754 2015 29 10458