Strategic Uncertainty and the Genetic Algorithm Adaptation
This paper studies the behavior of the genetic algorithm in coordination games with strategic uncertainty that are characterized by the multiplicity of equilibria. The main objectives are to examine whether the genetic algorithm adaptation leads to coordination and what the long run properties of the dynamics are. The genetic algorithm behavior is compared to the evidence from the experiments with human subjects. The adaptation of genetic algorithm players results in coordination on an equilibrium. Which equilibrium is selected depends on the group size. The analysis of the long-run properties of the dynamics show that, as a result of the continuing effects of the genetic drift, the genetic algorithm can reach any of the equilibria of the game regardless of the group size. Simulations with large group treatments spend most of the time in Pareto-inferior equilibria, while simulations with small group treatments spend most of the time in Pareto-superior equilibria of the game.
KeywordsGenetic Algorithm Nash Equilibrium Binary String Average Effort Evolutionary Stable Strategy
Unable to display preview. Download preview PDF.
- Cooper, R., D. De Jong, R. Forsythe, and Th. Ross, 1990, ‘Selection criteria in coordination games’, American Economic Review 80, 218–233.Google Scholar
- Holland, J.H., 1975, Adaptation in Natural and Artificial Systems, Ann Arbor, MI: The University of Michigan Press.Google Scholar
- Katz, M. and C. Shapiro, 1985, ‘Network externalities, competition and compatibility’, American Economic Review 75, 424–440.Google Scholar
- Roth, A., 1995, ‘Introduction to experimental economics’, in J. Kagel and A. Roth (Eds), Handbook of Experimental Economics, Princeton, NJ: Princeton University Press, pp. 3109.Google Scholar
- Sargent, T.J., 1993, Bounded Rationality in Macroeconomics, Oxford: Oxford University Press.Google Scholar
- Van Huyck, J., R Battalio, and R. Beil, 1990, ‘Tacit coordination games, strategic uncertainty, and coordination failure’, American Economic Review 80, 234–248.Google Scholar