A Convergence Criterion for Multiobjective Evolutionary Algorithms Based on Systematic Statistical Testing

  • Heike Trautmann
  • Uwe Ligges
  • Jörn Mehnen
  • Mike Preuss
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5199)

Abstract

A systematic approach for determining the generation number at which a specific Multi-Objective Evolutionary Algorithm (MOEA) has converged for a given optimization problem is introduced. Convergence is measured by the performance indicators Generational Distance, Spread and Hypervolume. The stochastic nature of the MOEA is taken into account by repeated runs per generation number which results in a highly robust procedure. For each generation number the MOEA is repeated a fixed number of times, and the Kolmogorow-Smirnov-Test is used in order to decide if a significant change in performance is gained in comparison to preceding generations. A comparison of different MOEAs on a problem with respect to necessary generation numbers becomes possible, and the understanding of the algorithm’s behaviour is supported by analysing the development of the indicator values. The procedure is illustrated by means of standard test problems.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Binh, T.T.: A Multiobjective Evolutionary Algorithm: The Study Cases. Technical Report, Institute for Automation and Communication, Berleben, Germany (1999)Google Scholar
  2. Collette, Y., Siarry, P.: Multiobjective Optimization, Principles and Case Studies. Springer, Berlin (2003)Google Scholar
  3. Deb, K., Jain, S.: Running Performance Metrics for Evolutionary Multi-Objective Optimisation. In: Proceedings of the Fourth Asia-Pacific Conference on Simulated Evolution and Learning (SEAL 2002), Singapore, pp. 13–20 (2002)Google Scholar
  4. Deb, K., Pratap, A., Agarwal, S.: A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 6(8), 182–197 (2002)CrossRefGoogle Scholar
  5. Deb, K.: Multi-Objective Optimization using Evolutionary Algorithms. Wiley, New York (2004)Google Scholar
  6. Fonseca, C.M., Fleming, P.J.: Multiobjective Genetic Algorithms Made Easy: Selection, Sharing, and Mating Restriction. In: Proceedings of the First International Conference on Genetic Algorithms in Engineering Systems: Innovations and Applications, Sheffield, UK, pp. 42–52 (1995)Google Scholar
  7. Hanne, T.: On the convergence of multiobjective evolutionary algorithms. European Journal of Operational Research 117(3), 553–564 (1999)CrossRefMATHGoogle Scholar
  8. Hoos, H.H., Stützle, T.: Stochastic Local Search: Foundations and Applications. Elsevier / Morgan Kaufmann, San Francisco (2004)Google Scholar
  9. Katzenbeisser, W., Hackl, P.: An alternative to the Kolmogorov-Smirnov two-sample Test. Communications in Statistics – Theory and Methods 15, 1163–1177 (1986)CrossRefMATHMathSciNetGoogle Scholar
  10. Knowles, J., Corne, D.: On Metrics for Comparing Nondominated Sets. In: Proc. of the IEEE Congress on Evolutionary Computation (CEC), Piscataway, New Jersey, pp. 711–716 (2002)Google Scholar
  11. Laumanns, M.: Analysis and Applications of Evolutionary Multiobjective Optimization Algorithms. PhD Thesis, Computer Engineering and Networks Laboratory, ETH Zurich, Switzerland (2003)Google Scholar
  12. Laumanns, M., Thiele, L., Deb, K., Zitzler, E.: Combining Convergence and Diversity in Evolutionary Multi-objective Optimization, Evolutionary Computation 10(3), 263-282 (2002)Google Scholar
  13. Miller, R.G.: Simultaneous Statistical Inference, 2nd edn. Series in Statistics. Springer, Berlin (1981)CrossRefMATHGoogle Scholar
  14. Rudolph, G., Agapie, A.: Convergence Properties of Some Multi-objective Evolutionary Algorithms. In: Proc. of the IEEE Congress on Evolutionary Computation (CEC), pp. 1010–1016 (2000)Google Scholar
  15. Rudenko, O., Schoenauer, M.: A Steady Performance Stopping Criterion for Pareto-based Evolutionary Algorithms. In: Proceedings of the 6th International Multi-Objective Programming and Goal Programming Conference, Hammamet (Tunesia) (2004)Google Scholar
  16. Rudolph, G.: Self-Adaptive Mutations Lead to Premature Convergence. IEEE Transactions on Evolutionary Computation 5(4), 410–414 (2001)CrossRefGoogle Scholar
  17. Sheskin, D.J.: Handbook of Parametric and Nonparametric Statistical Procedures, 2nd edn. Chapman & Hall, New York (2000)MATHGoogle Scholar
  18. Siegel, S., Castellan Jr., N.J.: Nonparametric statistics for the behavioral sciences, 2nd edn. McGraw-Hill, New York (1988)Google Scholar
  19. Van Veldhuizen, D.A., Lamont, G.B.: Multiobjective Evolutionary Algorithm Research: A History and Analysis. Dept. Elec. Comput. Eng., Graduate School of Eng., Air Force Inst. Technol., Wright-Patterson, TR 98 03 (1998)Google Scholar
  20. Van Veldhuizen, D.A., Lamont, G.B.: Evolutionary computation and convergence to a Pareto front. In: Proceedings of the Third Annual Conference on Genetic Programming, San Francisco, CA, pp. 221–228 (1998a)Google Scholar
  21. Wald, A., Wolfowitz, J.: On a test whether two samples are from the same population. Annals of Mathematical Statistics 11, 147–162 (1940)CrossRefMathSciNetGoogle Scholar
  22. Zitzler, E., Thiele, L.: Multiobjective Evolutionary Algorithms: A Comparative Case Study and the Strength Pareto Approach. IEEE Transactions on Evolutionary Computation 3(4), 257–271 (1999)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Heike Trautmann
    • 1
  • Uwe Ligges
    • 1
  • Jörn Mehnen
    • 2
  • Mike Preuss
    • 3
  1. 1.Department of StatisticsTechnische Universität DortmundGermany
  2. 2.Decision Engineering CentreCranfield UniversityUK
  3. 3.Chair of Algorithm EngineeringTechnische Universität DortmundGermany

Personalised recommendations