CaltechAUTHORS
  A Caltech Library Service

Self-tuning experience weighted attraction learning in games

Ho, Teck H. and Camerer, Colin F. and Chong, Juin-Kuan (2007) Self-tuning experience weighted attraction learning in games. Journal of Economic Theory, 133 (1). pp. 177-198. ISSN 0022-0531. doi:10.1016/j.jet.2005.12.008. https://resolver.caltech.edu/CaltechAUTHORS:20100921-154733619

[img]
Preview
PDF - Submitted Version
See Usage Policy.

372kB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20100921-154733619

Abstract

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.


Item Type:Article
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1016/j.jet.2005.12.008 DOIArticle
ORCID:
AuthorORCID
Ho, Teck H.0000-0001-5210-4977
Camerer, Colin F.0000-0003-4049-1871
Chong, Juin-Kuan0000-0002-5187-8652
Additional Information:© 2006 Elsevier. Received 4 November 2004; revised 19 December 2005. Available online 13 February 2006. Thanks to participants in the 2000 Southern Economics Association meetings, the Wharton School Decision Processes Workshop, the University of Pittsburgh, the Berkeley Marketing Workshop, the Nobel Symposium on Behavioral and Experimental Economics (December 2001) and C. Mónica Capra, David Cooper, Vince Crawford, Ido Erev, Guillaume Frechette, and anonymous referees for helpful comments.
Subject Keywords:learning; experience weighted attraction; quantal response equilibrium; fictitious play; reinforcement learning
Issue or Number:1
Classification Code:JEL classification codes: C72; C91
DOI:10.1016/j.jet.2005.12.008
Record Number:CaltechAUTHORS:20100921-154733619
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20100921-154733619
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:20082
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:22 Sep 2010 17:47
Last Modified:08 Nov 2021 23:57

Repository Staff Only: item control page