On Classifier Systems and Genetic Algorithms Playing Iterated Prisoner’s Dilemma

The paper proposes an experimental setup to compare the performance of genetic algorithms and classifier systems. In economic theory, the use of artificial adaptive agents as substitutes for the homo oeconomicus raises important methodological issues. While the reductionist approach grounded on Olympic rationality offers full rationality as the unique reference point for problem solving, weaker notions of rationality generate a variety of processes and outcomes of decision-making. The paper gives some suggestions on the sensitivity of the behaviour of agents to the algorithmic choice and to the codification of knowledge. Results show that in an iterated prisoner’s dilemma interesting behavioural patterns (such as strategies that perform better than the titfor- tat) emerge