On the complexity of the El Farol Bar game: a sensitivity analysis

详细信息    查看全文

• 作者：Shu-Heng Chen ; Umberto Gostoli
• 关键词：El Farol Bar problem ; Good ; society equilibrium ; Social preferences ; Imitation ; Mutation ; Inequity aversion ; Keeping ; up ; with ; the ; Joneses
• 刊名：Evolutionary Intelligence
• 出版年：2016
• 出版时间：December 2016
• 年：2016
• 卷：9
• 期：4
• 页码：113-123
• 全文大小：
• 刊物类别：Engineering
• 刊物主题：Appl.Mathematics/Computational Methods of Engineering; Artificial Intelligence (incl. Robotics); Statistical Physics and Dynamical Systems; Control, Robotics, Mechatronics; Bioinformatics; Application
• 出版者：Springer Berlin Heidelberg
• ISSN：1864-5917
• 卷排序：9

文摘

In this paper, we carry out a sensitivity analysis for an agent-based model of the use of public resources as manifested by the El Farol Bar problem. An early study using the same model has shown that a good-society equilibrium, characterized by both economic efficiency and economic equality, can be achieved probabilistically by a von Neumann network, and can be achieved surely with the presence of some agents having social preferences, such as the inequity-averse preference or the ‘keeping-up-with-the-Joneses’ preference. In this study, we examine this fundamental result by exploring the inherent complexity of the model; specifically, we address the effect of the three key parameters related to size, namely, the network size, the neighborhood size, and the memory size. We find that social preferences still play an important role over all the sizes considered. Nonetheless, it is also found that when network size becomes large, the parameter, the bar capacity (the attendance threshold), may also play a determining role.