Why Coevolution Doesn’t “Work”: Superiority and Progress in Coevolution
Coevolution often gives rise to counter-intuitive dynamics that defy our expectations. Here we suggest that much of the confusion surrounding coevolution results from imprecise notions of superiority and progress. In particular, we note that in the literature, three distinct notions of progress are implicitly lumped together: local progress (superior performance against current opponents), historical progress (superior performance against previous opponents) and global progress (superior performance against the entire opponent space). As a result, valid conditions for one type of progress are unduly assumed to lead to another. In particular, the confusion between historical and global progress is a case of a common error, namely using the training set as a test set. This error is prevalent among standard methods for coevolutionary analysis (CIAO, Master Tournament, Dominance Tournament, etc.) By clearly defining and distinguishing between different types of progress, we identify limitations with existing techniques and algorithms, address them, and generally facilitate discussion and understanding of coevolution. We conclude that the concepts proposed in this paper correspond to important aspects of the coevolutionary process.
KeywordsNash Equilibrium Solution Concept Local Progress Dominant Individual Pareto Dominance
Unable to display preview. Download preview PDF.
- 1.Axelrod, R.: The evolution of strategies in the iterated prisoner’s dilemma. In: Davis, L. (ed.) Genetic algorithms and simulated annealing, pp. 32–41. Morgan Kaufman, San Francisco (1987)Google Scholar
- 4.Rosin, C.D., Belew, R.K.: Methods for competitive co-evolution: Finding opponents worth beating. In: Eshelman, L. (ed.) Procs. 6th ICGA. Morgan Kaufmann, San Francisco (1995)Google Scholar
- 5.Ficici, S.G.: Solution Concepts in Coevolutionary Algorithms. Ph.D thesis, Brandeis University (May 2004)Google Scholar
- 6.Ficici, S.G.: Monotonic solution concepts in coevolution. In: Procs. GECCO 2005. ACM Press, New York (2005)Google Scholar
- 8.Stanley, K.O., Miikkulainen, R.: The dominance tournament method of monitoring progress in coevolution. In: Procs GECCO 2002 Workshops. Morgan Kaufmann, San Francisco (2002)Google Scholar
- 9.Ficici, S.G., Pollack, J.B.: Challenges in coevolutionary learning: arms-race dynamics, open-endedness, and medicocre stable states. In: Procs. ALIFE VI. MIT Press, Cambridge (1998)Google Scholar
- 11.Cartlidge, J.: Rules of Engagement: Competitive Coevolutionary Dynamics in Computational Systems. Ph.D thesis, The University of Leeds (2004)Google Scholar
- 13.Miconi, T.: The Road to Everywhere: Evolution, Complexity and Progress in Natural and Artificial Systems. Ph.D thesis, University of Birmingham (2007)Google Scholar
- 14.De Jong, E.D.: The MaxSolve algorithm for coevolution. In: Procs. GECCO 2005. ACM Press, New York (2005)Google Scholar