TY - GEN
T1 - Relational macros for transfer in reinforcement learning
AU - Torrey, Lisa
AU - Shavlik, Jude
AU - Walker, Trevor
AU - MacLin, Richard
PY - 2008
Y1 - 2008
N2 - We describe an application of inductive logic programming to transfer learning. Transfer learning is the use of knowledge learned in a source task to improve learning in a related target task. The tasks we work with are in reinforcement-learning domains. Our approach transfers relational macros, which are finite-state machines in which the transition conditions and the node actions are represented by first-order logical clauses. We use inductive logic programming to learn a macro that characterizes successful behavior in the source task, and then use the macro for decision-making in the early learning stages of the target task. Through experiments in the RoboCup simulated soccer domain, we show that Relational Macro Transfer via Demonstration (RMT-D) from a source task can provide a substantial head start in the target task.
AB - We describe an application of inductive logic programming to transfer learning. Transfer learning is the use of knowledge learned in a source task to improve learning in a related target task. The tasks we work with are in reinforcement-learning domains. Our approach transfers relational macros, which are finite-state machines in which the transition conditions and the node actions are represented by first-order logical clauses. We use inductive logic programming to learn a macro that characterizes successful behavior in the source task, and then use the macro for decision-making in the early learning stages of the target task. Through experiments in the RoboCup simulated soccer domain, we show that Relational Macro Transfer via Demonstration (RMT-D) from a source task can provide a substantial head start in the target task.
UR - http://www.scopus.com/inward/record.url?scp=40249114836&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=40249114836&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-78469-2_25
DO - 10.1007/978-3-540-78469-2_25
M3 - Conference contribution
AN - SCOPUS:40249114836
SN - 3540784683
SN - 9783540784685
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 254
EP - 268
BT - Inductive Logic Programming - 17th International Conference, ILP 2007, Revised Selected Papers
T2 - 17th International Conference on Inductive Logic Programming, ILP 2007
Y2 - 19 June 2007 through 21 June 2007
ER -