Characterizing Diversity in Genetic Programming

  • Bart Wyns
  • Peter De Bruyne
  • Luc Boullart
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3905)


In many evolutionary algorithms candidate solutions run the risk of getting stuck in local optima after a few generations of optimization. In this paper two improved approaches to measure population diversity are proposed and validated using two traditional test problems in genetic programming literature. Code growth gave rise to improve pseudo-isomorph measures by eliminating non-functional code using an expression simplifier. Also, Rosca’s entropy to measure behavioral diversity is updated to cope with problems producing a more continuous fitness value. Results show a relevant improvement with regard to the original diversity measures.


Genetic Programming Regression Problem Parse Tree Symbolic Regression Program Size 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Burke, E., Gustafson, S., Kendall, G.: A survey and analysis of diversity measures in genetic programming. In: Langdon, W.B., et al. (eds.) Proceedings of the Genetic and Evolutionary Computation Conference, pp. 716–723. Morgan Kaufmann, San Francisco (2002)Google Scholar
  2. 2.
    Rosca, J.P.: Genetic programming exploratory power and the discovery of functions. In: McDonnell, J.R., Reynolds, R.G., Fogel, D.B. (eds.) Evolutionary Programming IV Proceedings of the Fourth Annual Conference on Evolutionary Programming, pp. 719–736 (1995)Google Scholar
  3. 3.
    Rosca, J.P.: Entropy-driven adaptive representation. In: Rosca, J. (ed.) Proceedings of the Workshop on Genetic Programming: From Theory to Real-World Applications, Tahoe City, California, pp. 23–32 (1995)Google Scholar
  4. 4.
    Burke, E.K., Gustafson, S., Kendall, G.: Diversity in genetic programming: An analysis of measures and correlation with fitness. IEEE Transactions on Evolutionary Computation 8, 47–62 (2004)CrossRefGoogle Scholar
  5. 5.
    Soule, T., Foster, J.: Effects of code growth and parsimony pressure on populations in genetic programming. Evolutionary Computation 6, 293–309 (1998)CrossRefGoogle Scholar
  6. 6.
    Koza, J.R.: Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, Cambridge (1992)MATHGoogle Scholar
  7. 7.
    Punch, B., Zongker, D.: Lilgp 1.01 user’s manual. Technical report, Michigan State University, Michigan, USA (1996)Google Scholar
  8. 8.
    Gustafson, S., Ekárt, A., Burke, E.: Problem difficulty and code growth in genetic programming. Genetic programming and Evolvable Machines 5, 271–290 (2004)CrossRefGoogle Scholar
  9. 9.
    Wyns, B., Sette, S., Boullart, L.: Self-improvement to control code growth in genetic programming. In: Liardet, P., Collet, P., Fonlupt, C., Lutton, E., Schoenauer, M. (eds.) EA 2003. LNCS, vol. 2936, pp. 256–266. Springer, Heidelberg (2004)Google Scholar
  10. 10.
    Ekárt, A., Németh, S.: Selection based on the pareto nondominance criterion for controlling code growth in genetic programming. Genetic Programming and Evolvable Machines 2, 61–73 (2001)CrossRefMATHGoogle Scholar
  11. 11.
    de Jong, E., Pollack, J.: Multi-objective methods for tree size control. Genetic Programming and Evolvable Machines 4, 211–233 (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Bart Wyns
    • 1
  • Peter De Bruyne
    • 1
  • Luc Boullart
    • 1
  1. 1.Department of Electrical EnergySystem and Automation Ghent UniversityZwijnaardeBelgium

Personalised recommendations