Advertisement

Fitness causes bloat: Mutation

  • W. B. Langdon
  • R. Poli
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1391)

Abstract

The problem of evolving, using mutation, an artificial ant to follow the Santa Fe trail is used to study the well known genetic programming feature of growth in solution length. Known variously as “bloat”, “fluff” and increasing “structural complexity”, this is often described in terms of increasing “redundancy” in the code caused by “introns”.

Comparison between runs with and without fitness selection pressure, backed by Price’s Theorem, shows the tendency for solutions to grow in size is caused by fitness based selection. We argue that such growth is inherent in using a fixed evaluation function with a discrete but variable length representation. With simple static evaluation search converges to mainly finding trial solutions with the same fitness as existing trial solutions. In general variable length allows many more long representations of a given solution than short ones. Thus in search (without a length bias) we expect longer representations to occur more often and so representation length to tend to increase. I.e. fitness based selection leads to bloat.

Keywords

Genetic Programming Program Size Program Length Fitness Selection Parsimony Pressure 
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. [Alt95]
    Lee Altenberg. The Schema Theorem and Price’s Theorem. In L. Darrell Whitley and Michael D. Vose, editors, Foundations of Genetic Algorithms 3, pages 23–49, Estes Park, Colorado, USA, 31 July–2 August 1994 1995. Morgan Kaufmann.Google Scholar
  2. [Ang94]
    Peter John Angeline. Genetic programming and emergent intelligence. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 4, pages 75–98. MIT Press, 1994.Google Scholar
  3. [Bli96]
    Tobias Blickle. Theory of Evolutionary Algorithms and Application to System Synthesis. PhD thesis, Swiss Federal Institute of Technology, Zurich, November 1996.Google Scholar
  4. [BT94]
    Tobias Blickle and Lothar Thiele. Genetic programming and redundancy. In J. Hopf, editor, Genetic Algorithms within the Framework of Evolutionary Computation (Workshop at KI-94, Saarbrücken), pages 33–38, Im Stadtwald, Building 44, D-66123 Saarbrücken, Germany, 1994. Max-Planck-Institut für Informatik (MPI-I-94-241).Google Scholar
  5. [GR94]
    Chris Gathercole and Peter Ross. Dynamic training subset selection for supervised learning in genetic programming. In Yuval Davidor, Hans-Paul Schwefel, and Reinhard MÄnner, editors, Parallel Problem Solving from Nature III, pages 312–321, Jerusalem, 9–14 October 1994. Springer-Verlag.Google Scholar
  6. [GR96]
    Chris Gathercole and Peter Ross. An adverse interaction between crossover and restricted tree depth in genetic programming. In John R. Koza, David E. Goldberg, David B. Fogel, and Rick L. Riolo, editors, Genetic Programming 1996: Proceedings of the First Annual Conference, pages 291–296, Stanford University, CA, USA, 28–31 July 1996. MIT Press.Google Scholar
  7. [Hay96]
    Thomas Haynes. Duplication of coding segments in genetic programming. In Proceedings of the Thirteenth National Conference on Artificial Intelligence, pages 344–349, Portland, OR, August 1996.Google Scholar
  8. [IdS94]
    Hitoshi Iba, Hugo de Garis, and Taisuke Sato. Genetic programming using a minimum description length principle. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 12, pages 265–284. MIT Press, 1994.Google Scholar
  9. [Koz92]
    John R. Koza. Genetic Programming: On the Programming of Computers by Natural Selection. MIT Press, Cambridge, MA, USA, 1992.Google Scholar
  10. [Koz94]
    John R. Koza. Genetic Programming II: Automatic Discovery of Reusable Programs. MIT Press, Cambridge Massachusetts, May 1994.Google Scholar
  11. [Lan95]
    W. B. Langdon. Evolving data structures using genetic programming. In L. Eshelman, editor, Genetic Algorithms: Proceedings of the Sixth International Conference (ICGA95), pages 295–302, Pittsburgh, PA, USA, 15–19 July 1995. Morgan Kaufmann.Google Scholar
  12. [Lan96]
    William B. Langdon. Data structures and genetic programming. In Peter J. Angeline and K. E. Kinnear, Jr., editors, Advances in Genetic Programming 2, chapter 20, pages 395–414. MIT Press, Cambridge, MA, USA, 1996.Google Scholar
  13. [Lan98a]
    W. B. Langdon. Data Structures and Genetic Programming. Kulwer, 1998. Forthcoming.Google Scholar
  14. [Lan98b]
    W. B. Langdon. The evolution of size in variable length representations. In 1998 IEEE International Conference on Evolutionary Computation, Anchorage, Alaska, USA, 5–9 May 1998. Forthcoming.Google Scholar
  15. [LP97a]
    W. B. Langdon and R. Poli. An analysis of the MAX problem in genetic programming. In John R. Koza, Kalyanmoy Deb, Marco Dorigo, David B. Fogel, Max Garzon, Hitoshi Iba, and Rick L. Riolo, editors, Genetic Programming 1997: Proceedings of the Second Annual Conference, pages 222–230, Stanford University, CA, USA, 13–16 July 1997. Morgan Kaufmann.Google Scholar
  16. [LP97b]
    W. B. Langdon and R. Poli. Fitness causes bloat. In P. K. Chawdhry, R. Roy, and R. K. Pan, editors, Second On-line World Conference on Soft Computing in Engineering Design and Manufacturing. Springer-Verlag London, 23–27 June 1997.Google Scholar
  17. [LP97c]
    W. B. Langdon and R. Poli. Fitness causes bloat: Mutation. In John Koza, editor, Late Breaking Papers at the GP-97 Conference, pages 132–140, Stanford, CA, USA, 13–16 July 1997. Stanford Bookstore.Google Scholar
  18. [LP98a]
    W. B. Langdon and R. Poli. Genetic programming bloat with dynamic fitness. In W. Banzhaf, R. Poli, M. Schoenauer, and T. C. Fogarty, editors, Proceedings of the First European Workshop on Genetic Programming, LNCS, Paris, 14–15 April 1998. Springer-Verlag. Forthcoming.Google Scholar
  19. [LP98b]
    W. B. Langdon and R. Poli. Why ants are hard. Technical Report CSRP-98-4, University of Birmingham, School of Computer Science, January 1998. submitted to GP-98.Google Scholar
  20. [MM95]
    Nicholas Freitag McPhee and Justin Darwin Miller. Accurate replication in genetic programming. In L. Eshelman, editor, Genetic Algorithms: Proceedings of the Sixth International Conference (ICGA95), pages 303–309, Pittsburgh, PA, USA, 15–19 July 1995. Morgan Kaufmann.Google Scholar
  21. [NB95]
    Peter Nordin and Wolfgang Banzhaf. Complexity compression and evolution. In L. Eshelman, editor, Genetic Algorithms: Proceedings of the Sixth International Conference (ICGA95), pages 310–317, Pittsburgh, PA, USA, 15–19 July 1995. Morgan Kaufmann.Google Scholar
  22. [NFB95]
    Peter Nordin, Frank Francone, and Wolfgang Banzhaf. Explicitly defined introns and destructive crossover in genetic programming. In Justinian P. Rosca, editor, Proceedings of the Workshop on Genetic Programming: From Theory to Real-World Applications, pages 6–22, Tahoe City, California, USA, 9 July 1995.Google Scholar
  23. [NFB96]
    Peter Nordin, Frank Francone, and Wolfgang Banzhaf. Explicitly defined introns and destructive crossover in genetic programming. In Peter J. Angeline and K. E. Kinnear, Jr., editors, Advances in Genetic Programming 2, chapter 6, pages 111–134. MIT Press, Cambridge, MA, USA, 1996.Google Scholar
  24. [PL98]
    Riccardo Poli and William B Langdon. On the ability to search the space of programs of standard, one-point and uniform crossover in genetic programming. Technical Report CSRP-98-7, University of Birmingham, School of Computer Science, January 1998. submitted to GP-98.Google Scholar
  25. [Pri70]
    George R. Price. Selection and covariance. Nature, 227, August 1:520–521, 1970.MathSciNetCrossRefGoogle Scholar
  26. [RB96a]
    Justinian P. Rosca and Dana H. Ballard. Complexity drift in evolutionary computation with tree representations. Technical Report NRL5, University of Rochester, Computer Science Department, Rochester, NY, USA, December 1996.Google Scholar
  27. [RB96b]
    Justinian P. Rosca and Dana H. Ballard. Discovery of subroutines in genetic programming. In Peter J. Angeline and K. E. Kinnear, Jr., editors, Advances in Genetic Programming 2, chapter 9, pages 177–202. MIT Press, Cambridge, MA, USA, 1996.Google Scholar
  28. [SFD96]
    Terence Soule, James A. Foster, and John Dickinson. Code growth in genetic programming. In John R. Koza, David E. Goldberg, David B. Fogel, and Rick L. Riolo, editors, Genetic Programming 1996: Proceedings of the First Annual Conference, pages 215–223, Stanford University, CA, USA, 28–31 July 1996. MIT Press.Google Scholar
  29. [Sim93]
    K. Sims. Interactive evolution of equations for procedural models. The Visual Computer, 9:466–476, 1993.CrossRefGoogle Scholar
  30. [Tac93]
    Walter Alden Tackett. Genetic programming for feature discovery and image discrimination. In Stephanie Forrest, editor, Proceedings of the 5th International Conference on Genetic Algorithms, ICGA-93, pages 303–309, University of Illinois at Urbana-Champaign, 17–21 July 1993. Morgan Kaufmann.Google Scholar
  31. [Tac94]
    Walter Alden Tackett. Recombination, Selection, and the Genetic Construction of Computer Programs. PhD thesis, University of Southern California, Department of Electrical Engineering Systems, 1994.Google Scholar
  32. [Tac95]
    Walter Alden Tackett. Greedy recombination and genetic search on the space of computer programs. In L. Darrell Whitley and Michael D. Vose, editors, Foundations of Genetic Algorithms 3, pages 271–297, Estes Park, Colorado, USA, 31 July–2 August 1994 1995. Morgan Kaufmann.Google Scholar
  33. [WL96]
    Annie S. Wu and Robert K. Lindsay. A survey of intron research in genetics. In Hans-Michael Voigt, Werner Ebeling, Ingo Rechenberg, and Hans-Paul Schwefel, editors, Parallel Problem Solving From Nature IV. Proceedings of the International Conference on Evolutionary Computation, volume 1141 of LNCS, pages 101–110, Berlin, Germany, 22–26 September 1996. Springer-Verlag.Google Scholar
  34. [ZM93]
    Byoung-Tak Zhang and Heinz Mühlenbein. Evolving optimal neural networks using genetic algorithms with Occam’s razor. Complex Systems, 7:199–220, 1993.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • W. B. Langdon
    • 1
  • R. Poli
    • 1
  1. 1.School of Computer ScienceUniversity of BirminghamBirminghamUK

Personalised recommendations