Advertisement

Parameter Tuning of Real-Valued Crossover Operators for Statistics Preservation

  • Hiroshi Someya
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5361)

Abstract

Parameters of real-valued crossover operators have been often tuned under a constraint for preserving statistics of infinite parental population. For applications in actual scenes, in a previous study, an alternative constraint, called unbiased constraint, considering finiteness of the population has been derived. To clarify the wide applicability of the unbiased constraint, this paper presents two additional studies: (1) applying it to various crossover operators in higher dimensional search space, and (2) generalization of it for preserving statistics of overall population. Appropriateness of the parameter setting based on the unbiased constraint has been supported in discussion on robust search.

Keywords

Real-coded genetic algorithm functional specialization hypothesis statistics preservation parameter tuning 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Herrera, F., Lozano, M.: Special issue on real coded genetic algorithms: Operators, models and foundations. Soft Computing 9(4) (2005)Google Scholar
  2. 2.
    Kita, H., et al.: Multi-parental extension of the unimodal normal distribution crossover for real-coded genetic algorithms. In: Proc. of CEC 1999, pp. 1581–1587 (1999)Google Scholar
  3. 3.
    Someya, H.: Theoretical parameter value for appropriate population variance of the distribution of children in real-coded GA. In: Proc. of CEC 2008, pp. 2722–2729 (2008)Google Scholar
  4. 4.
    Kita, H., Yamamura, M.: A functional specialization hypothesis for designing genetic algorithms. In: Proc. of SMC 1999, pp. 579–584 (1999)Google Scholar
  5. 5.
    Lozano, J.A., et al. (eds.): Towards a New Evolutionary Computation: Advances in the Estimation of Distribution Algorithms. Springer, Heidelberg (2006)Google Scholar
  6. 6.
    Hoel, P.G.: Elementary Statistics, 4th edn. John Wiley & Sons, Inc., Chichester (1976)zbMATHGoogle Scholar
  7. 7.
    Kimura, S., et al.: An extension of UNDX based on guidelines for designing crossover operators: Proposition and evaluation of ENDX (in Japanese). Trans. of the Society of Instrument and Control Engineers 36(12), 1162–1171 (2000)CrossRefGoogle Scholar
  8. 8.
    Higuchi, T., et al.: Theoretical analysis of simplex crossover for real-coded genetic algorithms. In: Proc. of PPSN VI, pp. 365–374 (2000)Google Scholar
  9. 9.
    Ono, I., et al.: A genetic algorithm with characteristic preservation for function optimization. In: Proc. of IIZUKA 1996, pp. 511–514 (1996)Google Scholar
  10. 10.
    Someya, H., Yamamura, M.: A robust real-coded evolutionary algorithm with toroidal search space conversion. Soft Computing 9(4), 254–269 (2005)CrossRefzbMATHGoogle Scholar
  11. 11.
    Kita, H., et al.: Theoretical analysis of the unimodal normal distribution crossover for real-coded genetic algorithms. In: Proc. of 1998 International Conference on Evolutionary Computation, pp. 529–534 (1998)Google Scholar
  12. 12.
    Beyer, H.G., Deb, K.: On self-adaptive features in real-parameter evolutionary algorithms. IEEE Trans. on Evolutionary Computation 5(3), 250–270 (2001)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Hiroshi Someya
    • 1
  1. 1.The Institute of Statistical MathematicsTokyoJapan

Personalised recommendations