Abstract
In a multi-agent system, it becomes possible to solve a complicated problem by cooperative behavior with others. When people act in a group, as they are predicting the others’ action, estimating the others’ intention, and also making eye contact with others, they are realizing cooperative behavior efficiently. In the present paper, we try to introduce the concept of eye contact into a multi-agent system. In order to realize eye contact, we firstly define attention degrees both from self to the other and from the other to self. After that, we propose an action decision method that self agent makes easy to choose a target agent and to choose actions approaching to the agent using the attention degrees. Through computer simulation using a pursuit problem, we show that the agents making eye contact each other pursue preys by approaching each other. Simultaneously, we compare the proposed system with the standard Q-learning system and verify the usefulness of the proposed system.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Stone, P., Veloso, M.: Multiagent Systems: A Survey from a Machine Learning Perspective. Autonomous Robots 8(3), 345–383 (2000)
Kaelbling, L.P., Littman, M.L., Moore, A.P.: Reinforcement Learning: A Survey. Journal of Artificial Intelligence Research 4, 237–285 (1996)
Sutton, R. S., Barto, A. G.: Reinforcement Learning: An Introduction. MIT Press (1998)
Bratman, M.E.: Intention, Plans and Practical Reason. Harvard University Press (1987)
Nagayuki, Y., Ishii, S., Ito, M., Shimohara, K., Doya, K.: A Multi-Agent Reinforcement Learning Method with the Estimation of the Other Agent’s Actions. In: Proceedings of the Fifth International Symposium on Artificial Life and Robotics, vol. 1, pp. 255–259 (2000)
Nagayuki, Y., Ito, M.: Reinforcement Learning Method with the Inference of the Other Agent’s Policy for 2-Player Stochastic Games. Transactions on the Institute of Electronics, Information and Communication Engineers J86-D-I(11), 821–829 (2003) (in Japanese)
Watkins, C.J.C.H., Dayan, P.: Q-learning. Machine Learning 8(3-4), 279–292 (1992)
Yokoyama, A., Omori, T., Ishikawa, S., Okada, H.: Modeling of Action Decision Process Based on Intention Estimation. In: Proceedings of Joint 4th International Conference on Soft Computing and Intelligent Systems and 9th International Symposium on Advanced Intelligent Systems, vol. TH-F3-1 (2008)
Yokoyama, A., Omori, T.: Model Based Analysis of Action Decision Process in Collaborative Task Based on Intention Estimation. Transactions on the Institute of Electronics, Information and Communication Engineers J92-A(11), 734–742 (2009) (in Japanese)
Kobayashi, K., Kanehira, R., Kuremoto, T., Obayashi, M.: An Action Selection Method Based on Estimation of Other’s Intention in Time-Varying Multi-agent Environments. In: Lu, B.-L., Zhang, L., Kwok, J. (eds.) ICONIP 2011, Part III. LNCS, vol. 7064, pp. 76–85. Springer, Heidelberg (2011)
Hayashi, Y., Kojiri, T., Watanabe, T.: Focus Support Interface Based on Actions for Collaborative Learning. Neurocomputing 73(4-6), 669–675 (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kobayashi, K., Kurano, T., Kuremoto, T., Obayashi, M. (2012). Cooperative Behavior Acquisition in Multi-agent Reinforcement Learning System Using Attention Degree. In: Huang, T., Zeng, Z., Li, C., Leung, C.S. (eds) Neural Information Processing. ICONIP 2012. Lecture Notes in Computer Science, vol 7665. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-34487-9_65
Download citation
DOI: https://doi.org/10.1007/978-3-642-34487-9_65
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-34486-2
Online ISBN: 978-3-642-34487-9
eBook Packages: Computer ScienceComputer Science (R0)