Evolving New Strategies The Evolution of Strategies in the Iterated Prisoner ' s Dilemma

In complex environments, individuals are not fully able to analyze the situation and calculate their optimal strategy. 1 Instead they can be expected to adapt their strategy over time based upon what has been effective and what has not. One useful analogy to the adaptation process is biological evolution. In evolution, strategies that have been relatively effective in a population become more widespread, and strategies that have been less effective become less common in the population. Biological evolution has been highly successful has discovering complex and effective methods of adapting to very rich environmental situations. This is accomplished by differential reproduction of the more successful individuals. The evolutionary process also requires that successful characteristics be inherited through a genetic mechanism that allows some chance for new strategies to be discovered. One genetic mechanism allowing new strategies to be discovered is mutation. Another mechanism is crossover, whereby sexual reproduction takes some genetic material from one parent and some from the other. The mechanisms that have allowed biological evolution to be so good at adaptation have been employed in the field of artificial intelligence. The artificial intelligence technique is called the "genetic algorithm" (Holland, 1975). While other methods of

[1]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[2]  R. Axelrod.  More Effective Choice in the Prisoner's Dilemma , 1980 .

[3]  W. Hamilton.  Sex versus non-sex versus parasite , 1980 .

[4]  W. Hamilton,et al.  The evolution of cooperation. , 1984, Science.

[5]  R. Michod.  The Genetics of Altruism, Scott A. Boorman, Paul R. Levit. Academic Press, New York (1980), xx, +459. Price $29.50 , 1982 .

[6]  W. Hamilton,et al.  Heritable true fitness and bright birds: a role for parasites? , 1982, Science.

[7]  A. Rubinstein.  Finite automata play the repeated prisoner's dilemma , 1986 .

[8]  Avi Wigderson,et al.  On Play by Means of Computing Machines , 1986, TARK.

[9]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[10]  John H. Holland,et al.  Distributed genetic algorithms for function optimization , 1989 .

[11]  W. Hamilton,et al.  Sexual reproduction as an adaptation to resist parasites (a review). , 1990, Proceedings of the National Academy of Sciences of the United States of America.

[12]  L. Samuelson,et al.  Evolutionary stability in repeated games played by finite automata , 1992 .

[13]  R. Riolo.  Survival of the Fittest Bits , 1992 .

[14]  Dan Boneh,et al.  On genetic algorithms , 1995, COLT '95.

[15]  John H. Miller,et al.  The coevolution of automata in the repeated Prisoner's Dilemma , 1996 .