ABSTRACT
The number of offspring produced by each parent---that is, the fecundity of reproducing individuals---varies among evolutionary computation methods and settings. In most prior work fecundity has been tied directly to selectivity, with higher selection pressure giving rise to higher fecundity among individuals selected to reproduce. In nature, however, there is a wider variety of strategies, with different organisms producing different numbers of offspring under the influence of a range of factors including not only selection pressure but also other factors such as environmental stability and competition within a niche. In this work we consider possible lessons that may be drawn from nature's approaches to these issues and applied to evolutionary computation systems. In particular, we consider ways in which fecundity can be dissociated from selectivity and situations in which it may be beneficial to do so. We present a simple modification to the standard evolutionary algorithm, called decimation, that permits high fecundity in conjunction with modest selection pressure and which could be used in various forms of evolutionary computation. We also present a simple example, showing that decimation can improve the problem-solving performance of a genetic algorithm when applied to a deceptive problem.
- A. E. Eiben and J. E. Smith. Introduction to Evolutionary Computing. SpringerVerlag, 2003. Google ScholarDigital Library
- J. R. Koza. Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, Cambridge, MA, USA, 1992. Google ScholarDigital Library
- R. H. MacArthur and E. O. Wilson. The theory of island biogeography. Princeton University Press, April 2001.Google Scholar
Index Terms
- Fecundity and selectivity in evolutionary computation
Recommendations
Differential Evolution with Mutation Operators Based on Truncation
ICCIS '13: Proceedings of the 2013 International Conference on Computational and Information SciencesDifferential evolution(DE) is one of the most efficient algorithms in evolutionary algorithm family. Mutation operator is the core operator in DE. Generally, the parents in mutation operator are randomly selected from the parent population. In this paper,...
Fast non-elitist evolutionary algorithms with power-law ranking selection
GECCO '22: Proceedings of the Genetic and Evolutionary Computation ConferenceTheoretical evidence suggests that non-elitist evolutionary algorithms (EAs) with non-linear selection mechanisms can efficiently overcome broad classes of local optima where elitist EAs fail. However, the analysis assumes a weak selective pressure and ...
Ecological theory provides insights about evolutionary computation
GECCO '18: Proceedings of the Genetic and Evolutionary Computation Conference CompanionPromoting diversity in an evolving population is important for Evolutionary Computation (EC) because it reduces premature convergence on suboptimal fitness peaks while still encouraging both exploration and exploitation [3]. However, some types of ...
Comments