Advertisement

The Effect of Degenerate Coding on Genetic Algorithms

  • Colin R. Reeves
  • Ping Dai
Conference paper

Abstract

The choice of a suitable coding for the application of a GA to optimization problems is often critical to the effectiveness of the GA in finding good solutions. The phenomenon of degeneracy has been pointed out by Radcliffe as one that may be detrimental to GA performance [1, 2]. This problem is characteristic of applications of GAs to such cases as the travelling salesman problem (TSP), neural network design and training, and system identification in control. Previous experimental work by Hancock [3] found that the problem in practice appears less detrimental than expected. In this paper we examine a simple probability model for the occurrence of degenerate crossover that explains why degeneracy causes difficulties in some cases and not others. Experimental results in the case of system identification verify these expectations.

Keywords

Genetic Algorithm Travel Salesman Problem Travel Salesman Problem Recombination Operator Genetic Algorithm Performance 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    N.J. Radcliffe (1991) Equivalence class analysis of genetic algorithms. Complex Systems, 5, 183–205.MathSciNetMATHGoogle Scholar
  2. [2]
    N.J. Radcliffe (1993) Genetic set recombination and its application to neural network topology optimisation. Neural Computing and Applications, 1, 67–90.MATHCrossRefGoogle Scholar
  3. [3]
    P.J.B. Hancock (1992) Genetic algorithms and permutation problems: A comparison of recombination operators for neural net structure specification. In L.D. Whitley and J.D. Schaffer (Eds.) (1992) Proceedings of • COGANN-92: International Workshop on Combinations of Genetic Algorithms and Neural Networks, L.D.Whitley and J.D.Schaffer (eds.), IEEE Computer Society Press, Los Alamitos, CA, 108–122.Google Scholar
  4. [4]
    J.D. Schaffer, D. Whitley and L.J. Eshelman (1992) Combinations of genetic algorithms and neural networks: A survey of the state of the art. In L.D. Whitley and J.D. Schaffer (Eds.) (1992) Proceedings of COGANN-92: International Workshop on Combinations of Genetic Algorithms and Neural Networks, IEEE Computer Society Press, Los Alamitos, CA, 1–37.CrossRefGoogle Scholar
  5. [5]
    N.J. Radcliffe and P. Surry (1995) Formae and the variance of fitness. In D. Whitley and M. Vose (1995) Foundations of Genetic Algorithms 3, Morgan Kaufmann, San Mateo, CA, 51–72.Google Scholar
  6. [6]
    R. Nambiar and P. Mars (1992) Genetic algorithms for adaptive digital filtering. Proc. IEE Colloquium on Genetic Algorithms for Control and Systems Engineering. Digest No.1992/106, IEE, London.Google Scholar
  7. [7]
    Ping Dai (1997) Hybrid Genetic Algorithms with Application to System Identification and Control. MPhil dissertation, Control Theory and Applications Centre, Coventry University, UK.Google Scholar

Copyright information

© Springer-Verlag Wien 1999

Authors and Affiliations

  • Colin R. Reeves
    • 1
  • Ping Dai
    • 1
  1. 1.CTAC Computational Intelligence Group School of Mathematical and Information SciencesCoventry UniversityUK

Personalised recommendations