Abstract
This paper discusses a behavior of agents under multi-agent environment in game theory. Then we assume the behavior takes probabilistic Nash equilibrium in reinforcement learning. It is well-known that the behavior provides us with poor properties. For instance, no Nash equilibrium correspond to Pareto optimal and we can’t guarantee the convergence of learning. There, it is difficult to develop a multi-agent system to proceed cooperative work with agents. This paper takes the other approach to employee mixed Nash strategy based on correlated technique in terms of Local Effect Functions, and the model is useful to achieve cooperation among agents and they are designed to assess the convergence in learning through experiments in practice.
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
Aumann, R.J.: Subjectivity and correlation in randomized strategies. Journal of Mathematical Economics 1, 67–96 (1974)
Hu, J., Wellman, M.P.: Nash Q-Learning for General-Sum Stochastic Games. Journal of Machine Learning Research 4, 1039–1069 (2003)
Igoshi, K., Miura, T.: Strategic Knowledge By Nash-Q Learning for Reward Distribution. In: First IEEE International Conference on the Applications of Digital Information and Web Technologies, ICADIWT (2008)
Leyton-brown, K., Tennenholtz, M.: Local-Effect Games. In: International Joint Conference on Artificial Intelligence, IJCAI (2003)
Kok, J.R., Vlassis, N.: Collaborative Multiagent Reinforcement Learning by Payoff. Journal of Machine Learning Research 7, 1789–1828 (2006)
Rosenthal, R.W.: A class of games possessing pure-strategy Nash equilibrium. International Journal of Game Theory 2(1), 65–67 (1973)
Shoham, Y., Powers, R., Grenager, T.: Multi-Agent Reinforcement Learning - A Critical Survey, Technical Report (2003)
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (1998)
Watkins, C.J.C.H., Dayan, P.: Technical note: Q-Learning. Machine Learning 8, 279–292 (1992)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Igoshi, K., Miura, T., Shioya, I. (2010). Multi-agent System Environment Based on Repeated Local Effect Functions. In: Zavoral, F., Yaghob, J., Pichappan, P., El-Qawasmeh, E. (eds) Networked Digital Technologies. NDT 2010. Communications in Computer and Information Science, vol 88. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-14306-9_35
Download citation
DOI: https://doi.org/10.1007/978-3-642-14306-9_35
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-14305-2
Online ISBN: 978-3-642-14306-9
eBook Packages: Computer ScienceComputer Science (R0)