A generalized approach to belief learning in repeated games

• 成果类型: Article
• 署名作者: Ioannou, Christos A.; Romero, Julian
• 署名单位: University of Southampton; Purdue University System; Purdue University
• 刊物名称: GAMES AND ECONOMIC BEHAVIOR
• ISSN/ISSBN: 0899-8256
• DOI: 10.1016/j.geb.2014.05.007
• 发表日期: 2014
• 页码: 178-203
• 关键词: Adaptive models; Belief learning; Repeated-game strategies; Finite automata; prisoner's dilemma; Battle of the Sexes; Stag-Hunt; Chicken

摘要

We propose a methodology that is generalizable to a broad class of repeated games in order to facilitate operability of belief-learning models with repeated-game strategies. The methodology consists of (1) a generalized repeated-game strategy space, (2) a mapping between histories and repeated-game beliefs, and (3) asynchronous updating of repeated-game strategies. We implement the proposed methodology by building on three proven action-learning models. Their predictions with repeated-game strategies are then validated with data from experiments with human subjects in four, symmetric 2 x 2 games: Prisoner's Dilemma, Battle of the Sexes, Stag-Hunt, and Chicken. The models with repeated-game strategies approximate subjects' behavior substantially better than their respective models with action learning. Additionally, inferred rules of behavior in the experimental data overlap with the predicted rules of behavior. (C) 2014 Elsevier Inc. All rights reserved.