Advertisement

Some Considerations on the Reason for Bloat

  • W. Banzhaf
  • W. B. Langdon
Article

Abstract

A representation-less model for genetic programming is presented. The model is intended to examine the mechanisms that lead to bloat in genetic programming (GP). We discuss two hypotheses (“fitness causes bloat” and “neutral code is protective”) and perform simulations to examine the predictions deduced from these hypotheses. Our observation is that predictions from both hypotheses are realized in the simulated model.

genetic programming linear genomes effective fitness neutral variations 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    L. Altenberg, “Emergent phenomena in genetic programming,” in Proc. 3rd Annual Conf. Evolutionary Programming, A. Sebald and D. Fogel (eds.), World Scientific: Singapore, 1994, pp. 233–241.Google Scholar
  2. 2.
    L. Altenberg, “The evolution of evolvability in genetic programming,” in Advances in Genetic Programming, K. Kinnear Jr. (ed.), MIT Press: Cambridge, MA, 1994, pp. 47–74.Google Scholar
  3. 3.
    P. J. Angeline and J. Pollack, “Competitive environments evolve better solutions for complex tasks,” in Proc. Fifth Int. Conf. Genetic Algorithms (ICGA-93), Champaign-Urbana, IL. S. Forrest (ed.), Morgan Kaufmann: San Francisco, CA, 1993, pp. 264–270.Google Scholar
  4. 4.
    P. J. Angeline, “Genetic programming and emergent intelligence,” in Advances in Genetic Programming, K. Kinnear Jr. (ed.), MIT Press: Cambridge, MA, 1994, pp. 75–98.Google Scholar
  5. 5.
    W. Banzhaf, P. Nordin, R. Keller, and F. Francone, Genetic Programming–An Introduction, Morgan Kaufmann: San Francisco, CA, 1998.Google Scholar
  6. 6.
    T. Blickle and L. Thiele, “Genetic programming and redundancy,” in Genetic Algorithms within the Framework of Evolutionary Computation (Workshop at KI-94, Saarbrücken), Im Stadtwald, Building 44, D-66123 Saarbrücken, Germany, Max-Planck-Institut für Informatik Technical Report Nr. MPI-I-94-241, 1994, pp. 33–38.Google Scholar
  7. 7.
    M. Brameier and W. Banzhaf, “A comparison of linear genetic programming and neural networks in medical data mining,” IEEE Trans. Evolutionary Computation, vol. 5, pp. 17–26, 2001.Google Scholar
  8. 8.
    H. Iba, H. de Garis, and T. Sato, “Genetic programming using a minimum description length principle,” in Advances in Genetic Programming, K. E. Kinnear Jr. (ed.), MIT Press: Cambridge, MA, 1994, pp. 265–284.Google Scholar
  9. 9.
    J. R. Koza, Genetic Programming, MIT Press: Cambridge, MA, 1992.Google Scholar
  10. 10.
    W. B. Langdon, “The evolution of size in variable length representations,” in 1998 IEEE Int. Conf. Evolutionary Computation, Anchorage, Alaska, USA, 5–9 May 1998, IEEE Press: Piscataway, NJ, 1998, pp. 633–638.Google Scholar
  11. 11.
    W. B. Langdon, “Scaling of program tree fitness spaces,” Evolutionary Computation, vol. 7, no. 4, pp. 399–428, 1999.Google Scholar
  12. 12.
    W. B. Langdon and R. Poli, “Fitness causes bloat,” in Soft Computing in Engineering Design and Manufacturing. P. K. Chawdhry, R. Roy, and R. K. Pant (eds.), Springer-Verlag: London, 1997, pp. 13–22.Google Scholar
  13. 13.
    W. B. Langdon and R. Poli, “Fitness causes bloat: Mutation,” in Proc. First European Workshop on Genetic Programming, W. Banzhaf, R. Poli, M. Schoenauer and T. Fogarty (eds.), vol. 1391 of LNCS, Paris, 14–15 April 1998. Springer-Verlag: Berlin, 1998, pp. 37–48.Google Scholar
  14. 14.
    W. B. Langdon and R. Poli, “Genetic programming bloat with dynamic fitness,” in Proc. First European Workshop on Genetic Programming, W. Banzhaf, R. Poli, M. Schoenauer, and T. Fogarty (eds.), vol. 1391 of LNCS, Paris, 14–15 April 1998. Springer-Verlag: Berlin, 1998, pp. 96–112.Google Scholar
  15. 15.
    W. B. Langdon, T. Soule, R. Poli, and J. A. Foster, “The evolution of size and shape,” in Advances in Genetic Programming 3, L. Spector, W. B. Langdon, U.-M. O'Reilly and P. J. Angeline (eds.), MIT Press: Cambridge, MA, 1999, pp. 163–190.Google Scholar
  16. 16.
    W. B. Langdon, “Size fair and homologous tree crossover for tree GP,” Genetic Programming and Evolvable Machines, vol. 1, pp. 95–120, 2000.Google Scholar
  17. 17.
    W. B. Langdon and W. Banzhaf, “Genetic programming on difficult search spaces,” in Proc. Parallel Problem Solving from Nature, Paris 2000, M. Schoenauer et. al. (eds.), LNCS 1917, Springer-Verlag: Berlin, 2000, pp. 201–210.Google Scholar
  18. 18.
    W. B. Langdon and R. Poli, Foundations of Genetic Programming, Springer-Verlag: Berlin, 2002.Google Scholar
  19. 19.
    S. Luke, “Issues in Scaling Genetic Programming: Breeding Strategies, Tree Generation, and Bloat.” PhD thesis, University of Maryland, College Park, MD, 2000.Google Scholar
  20. 20.
    N. McPhee and J. Miller, “Accurate replication in genetic programming,” in Genetic Algorithms: Proc. Sixth Int. Conf. (ICGA95), Pittsburgh, PA, L. Eshelman (ed.), Morgan Kaufmann: San Francisco, CA, 1995, pp. 303–309.Google Scholar
  21. 21.
    P. Nordin and W. Banzhaf, “Complexity compression and evolution,” in Proc. Sixth Int. Conf. Genetic Algorithms (ICGA95), Pittsburgh, PA, L. Eshelman (ed.), Morgan Kaufmann: San Francisco, CA, 1995, pp. 310–317.Google Scholar
  22. 22.
    P. Nordin, W. Banzhaf, and F. Francone, “Introns in nature and in simulated structure evolution,” in Biocomputing and Emergent Computation, Proc. BCEC97 Skovde, Sweden, September 1–2, 1997, D. Lundh, B. Olsson, and A. Narayanan (eds.), World Scientific: Singapore, 1997, pp. 22–35.Google Scholar
  23. 23.
    R. Poli, “Exact schema theory for genetic programming and variable-length genetic algorithms with one-point crossover,” Genetic Programming and Evolvable Machines, vol. 2, pp. 123–163, 2001.Google Scholar
  24. 24.
    A. Singleton, Electronic Communication, 1994.Google Scholar
  25. 25.
    T. Soule, J. A. Foster, and J. Dickinson, “Code growth in genetic programming,” in Genetic Programming 1996: Proc. First Annual Conf., Stanford, CA, J. Koza, D. Goldberg, D. Fogel, and R. Riolo (eds.), MIT Press: Cambridge, MA, 1996, pp. 215–223.Google Scholar
  26. 26.
    T. Soule and J. A. Foster, “Code size and depth flows in genetic programming,” in Genetic Programming 1997: Proc. Second Annual Conf., Stanford, CA, J. R. Koza, K. Deb, M. Dorigo, D. B. Fogel, M. Garzon, H. Iba, and R. L. Riolo (eds.), Morgan Kaufmann: San Francisco, CA, 1997, pp. 313–320.Google Scholar
  27. 27.
    T. Soule, Code Growth in Genetic Programming. PhD thesis, University of Idaho, Moscow, ID, 1998.Google Scholar
  28. 28.
    C. Stephens and J. Mora Vargas, “Effective fitness as an alternative paradigm for evolutionary computation I: General formalism,” Genetic Programming and Evolvable Machines, vol. 1, pp. 363–378, 2000.Google Scholar
  29. 29.
    C. Stephens and J. Mora Vargas, “Effective fitness as an alternative paradigm for evolutionary computation II: Examples and applications,” Genetic Programming and Evolvable Machines, vol. 2, pp. 7–32, 2001.Google Scholar
  30. 30.
    W. A. Tackett, “Recombination, selection and the genetic construction of computer programs,” Dissertation: Department of Electrical Engineering Systems, University of Southern California, Los Angeles, CA, 1994.Google Scholar
  31. 31.
    B.-T. Zhang and H. Mühlenbein, “Genetic programming of minimal neural nets using Occam's razor,” in Proc. Fifth Int. Conf. Genetic Algorithms (ICGA-93), Champaign-Urbana, IL, S. Forrest (ed.), Morgan Kaufmann: San Francisco, CA, 1993, pp. 342–349.Google Scholar

Copyright information

© Kluwer Academic Publishers 2002

Authors and Affiliations

  • W. Banzhaf
    • 1
  • W. B. Langdon
    • 2
  1. 1.Department of Computer ScienceDortmund UniversityDortmundGermany
  2. 2.Computer ScienceUniversity College, LondonLondonUK

Personalised recommendations