Competitive evolution: A natural approach to operator selection

  • Q. Tuan Pham
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 956)


One of the main problems in applying evolutionary optimisation methods is the choice of operators and parameter values. This paper propose a competitive evolution method, in which several subpopulations are allowed to compete for computer time. The population with the fittest members, and that with the highest improvement rate in the recent past, are rewarded.

When using identical strategies in the subpopulations, this competitive strategy provides an insurance against unlucky runs while extracting only an insignificant cost in terms of extra function evaluations. When using different strategies in the subpopulations, it ensures that the best strategies are used and again the extra cost is not great. Competitive evolution is at its best when an operator — or the lack of it — may have a very detrimental effect which is not known in advance. Occasional mixing of the best performing subpopulations leads to further improvement.



normalised distance between two vectors x (equation 2)


function to be optimised


mutation frequency


base mutation frequency


population size


number of fitness evaluations


number of search variables


uniform crossovers/total crossovers


weighting factor indicating bias of offspring towards fitter parent


standard deviation of offspring from expected value, as a fraction of δ

x=(x1, x2...)

vector of search variables


minimum allowable value of xi


maximum allowable value of xi


distance for which the exponential part of the mutation frequency is 0.5


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Davis, L.: Adapting operator probabilities in genetic algorithms. Proc. 3rd Internat. Conf. on Genetic Algorithms, George Mason University, June 4, pp. 61–69. Morgan Kaufman Pub., San Mateo, California (1989)Google Scholar
  2. Davis, L., ed.: Handbook of Genetic Algorithms. Van Nostrand Reinhold, N.Y. (1991)Google Scholar
  3. De Jong, K.A.: An analysis of the behavior of a class of genetic adaptive systems, Doctoral thesis, Dept. of Computer and Communication Sciences, University of Michigan, Ann Arbor (1975)Google Scholar
  4. Fogel, D.B. and Atmar, J.W. (eds.): Proceedings of the First Annual Conference on Evolutionary Programming. Evolutionary Programming Society, San Diego, CA (1992)Google Scholar
  5. Goldberg, D.E.: Genetic Algorithms in Search, Optimization and Machine Learning, Addison-Wesley Pub, Reading, MA (1989)Google Scholar
  6. Grefenstette, J.J.: Optimization of control parameters for genetic algorithms. IEEE Trans. on Systems, Man and Cybernetics SMC-16 (1986) 122–8Google Scholar
  7. Holland, J.H.: Adaptation in natural and artificial systems. MIT Press (1975).Google Scholar
  8. Janikow, C.Z. and Michalewicz, Z.: An experimental comparison of binary and floating point representations in genetic algorithms.In R.K. Belew and L.B. Booker, editors, Proc. 4th Int. Conf. on Genetic Algorithms, pp.31–36. Morgan Kaufman (1991)Google Scholar
  9. Michalewicz, Z.: Genetic Algorithms + Data Structures = Evolution Programs. Springer Verlag, Berlin (1992)Google Scholar
  10. Rosenbrock, H.H.: An automatic method for finding the greatest or least value of a function. Computer J. 3 (1960) 175–184Google Scholar
  11. Rudolph, G.: Convergence analysis of canonical genetic algorithms. IEEE Trans on Neural Networks 5 (1994) 96–101Google Scholar
  12. Schaffer, J., David, R.A., Caruana, L.J. and Das, R.: A study of control parameters affecting online performance of genetic algorithms for function optimisation. In J.D. Schaffer (ed.), Proc. 3rd Internat. Conf. on Genetic Algorithms. Morgan Kaufman, San Mateo, California (1989)Google Scholar
  13. Shekel, J.: Test functions for multimodal search techniques. Fifth Annual Princeton Conf. Inform. Sci. Syst. (1971)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Q. Tuan Pham
    • 1
  1. 1.School of Chemical Engineering and Industrial ChemistryUniversity of New South WalesSydneyAustralia

Personalised recommendations