Advertisement

On the Utility of Redundant Encodings in Mutation-Based Evolutionary Search

  • Joshua D. Knowles
  • Richard A. Watson
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2439)

Abstract

A number of recent works in the evolutionary computation field have suggested that introducing large amounts of genetic redundancy may increase the evolvability of a population in an evolutionary algorithm. These works have variously claimed that the reliability of the search, the final fitness achieved, the ability to cope with changing environments, and the robustness to high mutation rates, may all be improved by employing this strategy. In this paper we dispute some of these claims, arguing that adding random redundancy cannot be generally useful for optimization purposes. By way of example we report on experiments where a proposed neutral encoding scheme (based on random Boolean networks) is compared to a direct encoding in two mutation-only EAs, at various mutation rates. Our findings show that with the appropriate choice of per-bit mutation rate, the evolvability of populations using the direct encoding is no less than with the redundant one.

Keywords

Mutation Rate Binary String High Mutation Rate Neutral Network Adaptive Landscape 
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.
    L. Barnett. Netcrawling—optimal evolutionary search with neutral networks. In Proceedings of the 2001 Congress on Evolutionary Computation (CEC2001). IEEE Service Centre, Piscataway NJ, 2001.Google Scholar
  2. 2.
    S. Bornholdt and K. Sneppen. Robustness as an evolutionary principle. Proceedings of the Royal Society of London Series B-Biological Sciences, 267:2281–2286, 2000.Google Scholar
  3. 3.
    S. Bullock. Smooth operator? Understanding and visualising mutation bias. In J. Kelemen and P. Sosik, editors, Sixth European Conference on Artificial Life, pages 602–612, 2001.Google Scholar
  4. 4.
    M. Ebner, M. Shackleton, and R. Shipman. How neutral networks influence evolvability. Complexity, 2002. (In press).Google Scholar
  5. 5.
    I. Harvey and A. Thompson. Through the labyrinth evolution finds a way: A silicon ridge. In Proceedings of ICES96, pages 406–422. Springer-Verlag, 1996.Google Scholar
  6. 6.
    S. A. Kauffman. Adaptation on rugged fitness landscapes. In D. Stein, editor, Lectures in the Sciences of Complexity, pages 527–618. Redwood City: Addison-Wesley, 1989. SFI Studies in the Sciences of Complexity, Lecture Volume I.Google Scholar
  7. 7.
    M. Kimura. The Neutral Theory of Molecular Evolution. Cambridge University Press, 1983.Google Scholar
  8. 8.
    M. Motoki and R. Uehara. Unique solution instance generation for the 3-satisfiability (3-SAT) problem. Technical Report C-129, Dept. of Math and Comp. Sciences, Tokyo Institute of Technology, 1999.Google Scholar
  9. 9.
    M. A. Potter. An NK-landscape generator, 1997. Presented at the Workshop on Test Problem Generators for Evolutionary Algorithms, at the 7th ICGA.Google Scholar
  10. 10.
    C. M. Reidys and P. F. Stadler. Neutrality in fitness landscapes. Appl. Math. Comput., 117:321–350, 1998.CrossRefMathSciNetGoogle Scholar
  11. 11.
    M. Shackleton, R. Shipman, and M. Ebner. An investigation of redundant genotype-phenotype mappings and their role in evolutionary search. In Proceedings of the 2000 Congress on Evolutionary Computation, pages 493–500, Piscataway, NJ, 2000. IEEE Service Center.Google Scholar
  12. 12.
    R. Shipman. Genetic redundancy: Desirable or problematic for evolutionary adaptation? In Proceedings of the 4th International Conference on Artificial Neural Networks and Genetic Algorithms (ICANNGA). Springer-Verlag, 1999.Google Scholar
  13. 13.
    R. Shipman, M. Shackleton, M. Ebner, and R. Watson. Neutral search spaces for artificial evolution: a lesson from life. In Artificial Life VII: Proceedings of the Seventh International Conference. MIT Press, 2000.Google Scholar
  14. 14.
    E. van Nimwegen and J. P. Crutchfield. Metastable evolutionary dynamics: Crossing fitness barriers or escaping via neutral paths? Bulletin of Mathematical Biology,62(5):799–848, Sep 2000.CrossRefGoogle Scholar
  15. 15.
    V. K. Vassilev and J. F. Miller. The advantages of landscape neutrality in digital circuit evolution. In ICES, pages 252–263, 2000.Google Scholar
  16. 16.
    R. A. Watson. Analysis of recombinative algorithms on a non-separable buildingblock problem. In Foundations of Genetic Algorithms VI, pages 69–89. Morgan Kaufmann, 2001.Google Scholar
  17. 17.
    R. A. Watson, G. S. Hornby, and J. B. Pollack. Modeling building-block interdependency. In Parallel Problem Solving from Nature— PPSN V, pages 97–106. Springer-Verlag, 1998.Google Scholar
  18. 18.
    T. Yu and J. Miller. Neutrality and the evolvability of Boolean function landscape. In Proceedings of the 4th European Conference on Genetic Programming (EuroGP), volume LNCS 2038, page 204 ff. Springer-Verlag, 2001.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Joshua D. Knowles
    • 1
  • Richard A. Watson
    • 2
  1. 1.IRIDIA— CP 194/6Université Libre de BruxellesBrusselsBelgium
  2. 2.DEMO, Volen Center for Complex SystemsBrandeis UniversityWalthamUSA

Personalised recommendations