Cold Play: Learning Across Bimatrix Games

Lensberg, Terje; Schenk-Hoppé, Klaus Reiner

doi:10.2139/ssrn.3432903

Cold Play: Learning Across Bimatrix Games

57 Pages Posted: 8 Aug 2019 Last revised: 30 Nov 2020

See all articles by Terje Lensberg

Terje Lensberg

Norwegian School of Economics (NHH) - Department of Finance

Klaus Reiner Schenk-Hoppé

University of Manchester - Department of Economics; Norwegian School of Economics (NHH) - Department of Finance

Date Written: March 10, 2020

Abstract

We study one-shot play in the set of all bimatrix games by a large population of agents. The agents never see the same game twice, but they can learn `across games' by developing solution concepts that tell them how to play new games. Each agent's individual solution concept is represented by a computer program, and natural selection is applied to derive a stochastically stable solution concept. Our aim is to develop a theory predicting how experienced agents would play in one-shot games. To use the theory, visit https://gplab.nhh.no/gamesolver.php.

Keywords: One-shot games, solution concepts, genetic programming, evolutionary stability

JEL Classification: C63, C73, C90

Suggested Citation: Suggested Citation

Lensberg, Terje and Schenk-Hoppé, Klaus Reiner, Cold Play: Learning Across Bimatrix Games (March 10, 2020). Available at SSRN: https://ssrn.com/abstract=3432903 or http://dx.doi.org/10.2139/ssrn.3432903