Learning correlated equilibria in normal form games
Learning correlated equilibria in normal form games
We analyze a population game as being constituted by a set of players, a normal form game and an interaction pattern. The latter specifies the way players are repeatedly matched in the population to play one shot of the normal form game.
We first relate the set of equilibria of the populations game to the set of correlated equilibria of the underlying game, and then focus on learning processes that we model as Markovian adaptive dynamics. For the class of doubly symmetric games, we formulate general conditions under which convergence is obtained under myopic best-reply dynamics. We also analyze noisy best-reply dynamics, where players' behaviour is perturbed by payoff dependent mistakes, and explicitly characterize the ergodic distribution of the population game in terms of the correlated equilibrium payoffs of the underlying game.
We conclude with ome good examples
University of Southampton
Ianni, A.
35024f65-34cd-4e20-9b2a-554600d739f3
January 1997
Ianni, A.
35024f65-34cd-4e20-9b2a-554600d739f3
Ianni, A.
(1997)
Learning correlated equilibria in normal form games
(Discussion Papers in Economics and Econometrics, 9713)
Southampton, GB.
University of Southampton
Record type:
Monograph
(Discussion Paper)
Abstract
We analyze a population game as being constituted by a set of players, a normal form game and an interaction pattern. The latter specifies the way players are repeatedly matched in the population to play one shot of the normal form game.
We first relate the set of equilibria of the populations game to the set of correlated equilibria of the underlying game, and then focus on learning processes that we model as Markovian adaptive dynamics. For the class of doubly symmetric games, we formulate general conditions under which convergence is obtained under myopic best-reply dynamics. We also analyze noisy best-reply dynamics, where players' behaviour is perturbed by payoff dependent mistakes, and explicitly characterize the ergodic distribution of the population game in terms of the correlated equilibrium payoffs of the underlying game.
We conclude with ome good examples
This record has no associated files available for download.
More information
Published date: January 1997
Identifiers
Local EPrints ID: 33187
URI: http://eprints.soton.ac.uk/id/eprint/33187
PURE UUID: d7f4572f-a866-401f-bcb1-d875a1b2f170
Catalogue record
Date deposited: 25 Jan 2008
Last modified: 12 Dec 2021 02:54
Export record
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
