Self-tuning experience weighted attraction learning in games

被引:113
作者
Ho, Teck H. [1 ]
Camerer, Colin F.
Chong, Juin-Kuan
机构
[1] Univ Calif Berkeley, Haas Sch Business, Berkeley, CA 94720 USA
[2] CALTECH, Div Humanities & Social Sci, Pasadena, CA 91125 USA
[3] Natl Univ Singapore, Singapore 117592, Singapore
[4] Sungkyunkwan Univ, Seoul 110745, South Korea
关键词
learning; experience weighted attraction; quantal response equilibrium; fictitious play; reinforcement learning;
D O I
10.1016/j.jet.2005.12.008
中图分类号
F [经济];
学科分类号
02 ;
摘要
Self-tuning experience weighted attraction (EWA) is a one-parameter theory of learning in games. It addresses a criticism that an earlier model (EWA) has too many parameters, by fixing some parameters at plausible values and replacing others with functions of experience so that they no longer need to be estimated. Consequently, it is econometrically simpler than the popular weighted fictitious play and reinforcement learning models. The functions of experience which replace free parameters "self-tune" over time, adjusting in a way that selects a sensible learning rule to capture subjects' choice dynamics. For instance, the self-tuning EWA model can turn from a weighted fictitious play into an averaging reinforcement learning as subjects equilibrate and learn to ignore inferior foregone payoffs. The theory was tested on seven different Games, and compared to the earlier parametric EWA model and a one-parameter stochastic equilibrium theory (QRE). Self-tuning EWA does as well as EWA in predicting behavior in new games, even though it has fewer parameters, and fits reliably better than the QRE equilibrium benchmark. (c) 2006 Elsevier Inc. All rights reserved.
引用
收藏
页码:177 / 198
页数:22
相关论文
共 53 条
[1]  
AMALDOSS W, 2005, CHOOSING RIGHT POND
[2]  
ARIFOVIC J, 2002, UNPUB COMPUTER TESTB
[3]  
ARTHUR WB, 1991, AM ECON REV, V81, P353
[4]   The winner's curse, reserve prices, and endogenous entry:: empirical insights from eBay auctions [J].
Bajari, P ;
Hortaçsu, A .
RAND JOURNAL OF ECONOMICS, 2003, 34 (02) :329-355
[5]  
Bush RR, 1955, Stochastic models for learning, DOI DOI 10.1037/14496-000
[6]  
CABRALAS A, 2002, EQUILIBRIUM SELECTIO
[7]   Experience-weighted attraction learning in normal form games [J].
Camerer, C ;
Ho, TH .
ECONOMETRICA, 1999, 67 (04) :827-874
[8]   Models of thinking, learning, and teaching in games [J].
Camerer, C ;
Ho, T ;
Chong, K .
AMERICAN ECONOMIC REVIEW, 2003, 93 (02) :192-195
[9]   A cognitive hierarchy model of games [J].
Camerer, CF ;
Ho, TH ;
Chong, JK .
QUARTERLY JOURNAL OF ECONOMICS, 2004, 119 (03) :861-898
[10]   Sophisticated experience-weighted attraction learning and strategic teaching in repeated games [J].
Camerer, CF ;
Ho, TH ;
Chong, JK .
JOURNAL OF ECONOMIC THEORY, 2002, 104 (01) :137-188