Existence of Risk Strategy Equilibrium in Games Having No Pure Strategy Nash Equilibrium
Two key properties defining an intelligent agent are reactive and pro-active. Before designing an intelligent agent for any multi-agent system, we need to first understand how agents should behave and interact in that particular application, which can be done by modelling the application as a game. To analyze these games and to understand how decision-makers interact, we can use a collection of analytical tools known as Game Theory. Risk strategies is a new kind of game-theoretic strategy. Simulations in previous work have shown that agents using risk strategies are reactive as well as pro-active and thus have better performance than agents using other models or strategies in various applications. However, research on risk strategies has been focusing on formalization, application, and games having pure strategy Nash equilibrium. In this paper, we analyze a game having no pure strategy Nash equilibrium. We find that risk strategy equilibrium may exist even the game does not have pure strategy Nash equilibrium. We then summarize general conditions for the existence of risk strategy equilibrium. Simulation shows that agents using risk strategies also have better performance than agents using other existing strategies in a game having no pure strategy Nash equilibrium.
KeywordsMultiagent System Pure Strategy Intelligent Agent Risk Attitude Pareto Optimum
Unable to display preview. Download preview PDF.
- 4.The Prisoner’s Dilemma Competition (2005), http://www.prisoners-dilemma.com/
- 5.Glass, A., Grosz, B.: Socially conscious decision-making. In: Proceedings of the Fourth International Conference on Autonomous Agents, pp. 217–224 (2000)Google Scholar
- 7.Lam, K.M., Leung, H.F.: Behavioral predictors and adaptive strategy for minority games. Physica A (submitted)Google Scholar
- 8.Lam, K.M., Leung, H.F.: Risk strategies and risk strategy equilibrium in agent interactions modeled as normal repeated 2 by 2 risk games. In: The Eighth Pacific Rim International Workshop on Multi-Agents (2005)Google Scholar
- 12.Lam, K.M., Leung, H.F.: An adaptive strategy for minority games. In: The 6th International Joint Conference on Autonomous Agents and Multiagent Systems (2007)Google Scholar
- 13.Lam, K.M., Leung, H.F.: An adaptive strategy for resource allocation modeled as minority game. In: The First IEEE International Conference on Self-Adaptive and Self-Organizing Systems, pp. 193–202 (2007)Google Scholar
- 14.Lam, K.M., Leung, H.F.: Incorporating risk attitude and reputation into infinitely repeated games and an analysis on the iterated prisoner’s dilemma. In: The 19th IEEE International Conference on Tools with Artificial Intelligence (2007)Google Scholar
- 15.Mui, L., Mohtashemi, M., Halberstadt, A.: A computational model of trust and reputation. In: Proceedings of 35th Hawaii International Conference on System Science (2002)Google Scholar
- 17.Rubiera, J.C., Lopez, J.M.M., Muro, J.D.: A fuzzy model of reputation in multi-agent systems. In: Proceedings of the Fifth International Conference on Autonomous Agents, pp. 25–26 (2001)Google Scholar
- 19.Sartain, A.Q., North, A.J., Strange, J.R., Chapman, H.M.: Psychology – Understanding Human Behavior. McGraw-Hill Book Company, New York (1962)Google Scholar