Genetic Algorithms in Time-Dependent Environments
The influence of time-dependent fitnesses on the infinite population dynamics of simple genetic algorithms (GAs) without crossover is analyzed. Based on general arguments, a schematic phase diagram is constructed that allows one to characterize the asymptotic states in dependence on the mutation rate and the time scale of changes. Furthermore, the notion of regular changes is raised for which the population can be shown to converge towards a generalized quasispecies. Based on this, error thresholds and an optimal mutation rate are approximately calculated for a generational GA with a moving needle-inthe-haystack landscape. The phase diagram thus found is fully consistent with our general considerations.
KeywordsTime-dependent fitness landscape quasispecies error threshold optimal mutation rate
Unable to display preview. Download preview PDF.
- 1.E. Baake and W. Gabriel. Biological evolution through mutation, selection, and drift: an introductory review. Ann. Rev. Comp. Phys., 7:203–264, 2000.Google Scholar
- 4.J. Branke. Evolutionary algorithms for dynamic optimization problems, a survey. Technical Report 387, AIFB University Karlsruhe, 1999.Google Scholar
- 5.J. Branke, M. Cutaia and H. Dold. Reducing genetic drift in steady state evolutionary algorithms. In W. Banzhaf et al., editors, Proceedings of GECCO1999, pages 68–74. Morgan Kaufmann, San Francisco, 1999.Google Scholar
- 6.K. DeJong and J. Sarma. Generation gaps revisited. In L. D. Whitley, editor, Foundations of Genetic Algorithms2, pages 19–28. Morgan Kaufmann, San Francisco, 1993.Google Scholar
- 8.M. Eigen and P. Schuster. The Hypercycle — a Principle of Natural Self-Organization. Springer-Verlag, Berlin Heidelberg New York, 1979.Google Scholar
- 10.M. Nilsson and N. Snoad. Error thresholds on dynamic fitness-landscapes. Working Paper 99–04–030, Santa Fe Institute, 1999.Google Scholar
- 11.A. Rogers and A. Prügel-Bennett. Modeling the dynamics of a steady state genetic algorithm. In W. Banzhaf and C. Reeves, editors, Foundations of Genetic Algorithms5, pages 57–68. Morgan Kaufmann, San Francisco, 1998.Google Scholar
- 12.J. E. Rowe. Finding attractors for periodic fitness functions. In W. Banzhaf et al., editors, Proceedings of the Genetic and Evolutionary Computation Conference, GECCO1999, pages 557–563. Morgan Kaufmann, San Francisco, 1999.Google Scholar
- 13.J. E. Rowe. The dynamical systems model of the simple genetic algorithm. This volume, pages 31–57.Google Scholar
- 14.J. E. Rowe. Cyclic attractors and quasispecies adaptability. This volume, pages 251— 259.Google Scholar
- 18.C. O. Wilke, C. Ronnewinkel and T. Martinetz. Molecular evolution in timedependent environments. In D. Floreano, J.-D. Nicoud and F. Mondada, editors, Proceedings of the European Conference on Artificial Life1999, pages 417–421. Springer-Verlag, Berlin Heidelberg New York, 1999.Google Scholar
- 19.C. O. Wilke and C. Ronnewinkel. Dynamic fitness landscapes in the quasispecies model. Physics Reports, Elsevier, in press.Google Scholar