Goal Evolution based on Adaptive Q-learning for Intelligent Agent

This paper presents an adaptive approach to address the goal evolution of the intelligent agent. When agents are initially created, they have some goals and few capabilities. These capabilities can perform some actions to satisfy their goals. They strive to adapt themselves to the low capabilities. Reinforcement learning method is used to the evolution of agent goal. An abstract agent programming language (3APL) is introduced to build the agent mental states. We propose reinforcement learning to refine the top-level goals. A robot soccer game is used to explain our approach. Moreover, we show how a refinement of the soccer player's mental state is derived from the evolving goals by reinforcement learning.

[1]  Shuhei Kinoshita,et al.  Team 11monkeys Description , 1999 .

[2]  Nicholas R. Jennings,et al.  Agent Theories, Architectures, and Languages: A Survey , 1995, ECAI Workshop on Agent Theories, Architectures, and Languages.

[3]  Erol Gelenbe,et al.  Simulation with learning agents , 2001, Proc. IEEE.

[4]  Frank Dignum,et al.  Issues in multiagent system development , 2004, Proceedings of the Third International Joint Conference on Autonomous Agents and Multiagent Systems, 2004. AAMAS 2004..

[5]  Shuhei Kinoshita,et al.  Team 11 monkeys Description 11 monkeys , 1999 .

[6]  Yoichiro Maeda.  Modified Q-learning method with fuzzy state division and adaptive rewards , 2002, 2002 IEEE World Congress on Computational Intelligence. 2002 IEEE International Conference on Fuzzy Systems. FUZZ-IEEE'02. Proceedings (Cat. No.02CH37291).

[7]  Frank Dignum,et al.  Autonomy and Agent Deliberation , 2003, Agents and Computational Autonomy.

[8]  Jonathan Lee,et al.  Evolutionary agents for intelligent transport systems , 2005 .

[9]  Andrea Bonarini.  Evolutionary learning, reinforcement learning, and fuzzy rules for knowledge acquisition in agent-based systems , 2001 .

[10]  Hisao Ishibuchi,et al.  An evolutionary approach for strategy learning in RoboCup soccer , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[11]  Daniel Kudenko,et al.  Learning in multi-agent systems , 2001, The Knowledge Engineering Review.

[12]  Frank Dignum,et al.  A Programming Language for Cognitive Agents Goal Directed 3APL , 2003, PROMAS.

[13]  Tomohiro Yamaguchi,et al.  Interactive multiagent reinforcement learning with motivation rules , 2001, Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001.

[14]  C.J.C.H. Watkins.  Automatic learning of efficient behaviour , 1989 .

[15]  Frank Dignum,et al.  Dynamics of Declarative Goals in Agent Programming , 2004, DALT.

[16]  Carlos D. Castillo,et al.  Chimps: an evolutionary reinforcement learning approach for soccer agents , 2003, SMC'03 Conference Proceedings. 2003 IEEE International Conference on Systems, Man and Cybernetics. Conference Theme - System Security and Assurance (Cat. No.03CH37483).

[17]  Kao-Shing Hwang,et al.  Cooperative strategy based on adaptive Q-learning for robot soccer systems , 2004, IEEE Transactions on Fuzzy Systems.

[18]  Jong-Yih Kuo,et al.  A document-driven agent-based approach for business processes management , 2004, Inf. Softw. Technol..

[19]  Michael Wooldridge,et al.  Reasoning about Visibility, Perception and Knowledge , 1999, ATAL.

[20]  Gregory M. P. O'Hare,et al.  Decision-making of BDI agents, a fuzzy approach , 2004, The Fourth International Conference onComputer and Information Technology, 2004. CIT '04..

[21]  Koen V. Hindriks,et al.  A Formal Architecture for the 3APL Agent Programming Language , 2000, ZB.