Advertisement

On the optimization of unimodal functions with the (1+1) evolutionary algorithm

  • Stefan Droste
  • Thomas Jansen
  • Ingo Wegener
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1498)

Abstract

We investigate the expected running time of the (1+1) EA, a very simple Evolutionary Algorithm, on the class of unimodal fitness functions with Boolean inputs. We analyze the behavior on a generalized version of long paths [6, 10] and prove an exponential lower bound on the expected running time. Thereby we show that unimodal functions can be very difficult to be optimized for the (1+1) EA. Furthermore, we prove that a little modification in the selection method can lead to huge changes in the expected running time.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Droste, S., Jansen, Th., and Wegener, I. (1998). A rigorous complexity analysis of the (1+1) Evolutionary Algorithm for linear functions with Boolean inputs. In Proceedings of the IEEE Congress on Evolutionary Computation (ICEC'98), 499–504. IEEE Press, Piscataway, NJ.Google Scholar
  2. 2.
    Fogel, D.B. (1995). Evolutionary Computation: Toward a New Philosophy of Machine Intelligence. IEEE Press, Piscataway, NJ.Google Scholar
  3. 3.
    Goldberg, D.E. (1989). Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley, Reading, Mass.Google Scholar
  4. 4.
    Höhn, C. and Reeves, C. (1996). Are long path problems hard for Genetic Algorithms? In Voigt, H.-M., Ebelin, W., Rechenberg, I., and Schwefel, H.-P. (Eds.): Parallel Problem Solving from Nature (PPSN IV), 134–143. Springer, Berlin. LNCS 1141.Google Scholar
  5. 5.
    Holland, J.H. (1975). Adaption in Natural and Artificial Systems. University of Michigan, Michigan.Google Scholar
  6. 6.
    Horn, J., Goldberg, D.E., and Deb, K. (1994). Long path problems. In Davidor, Y., Schwefel, H.-P., and Männer, R. (Eds.): Parallel Problem Solving From Nature (PPSN III), 149–158. Springer, Berlin. LNCS 866.Google Scholar
  7. 7.
    Koza, J.R. (1992). Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, Cambridge, Mass.Google Scholar
  8. 8.
    Mühlenbein, H. (1992). How Genetic Algorithms really work. Mutation and hillclimbing. In Männer, R. and Manderick, R. (Eds.), Parallel Problem Solving from Nature (PPSN II), 15–25. North-Holland, Amsterdam.Google Scholar
  9. 9.
    Rechenberg, I. (1994). Evolutionsstrategie '94. Frommann-Holzboog, Stuttgart.Google Scholar
  10. 10.
    Rudolph, G. (1997). Convergence Properties of Evolutionary Algorithms. Ph.D. Thesis. Verlag Dr. Kovač, Hamburg.Google Scholar
  11. 11.
    Rudolph, G. (1997b). How mutation and selection solve long-path problems in polynomial expected time. Evolutionary Computation 4(2), 195–205.MathSciNetGoogle Scholar
  12. 12.
    Schwefel, H.-P. (1995). Evolution and Optimum Seeking. Wiley, New York.Google Scholar
  13. 13.
    Schwefel, H.-P. (1997). Personal communication.Google Scholar
  14. 14.
    Wolpert, D.H. and Macready, W. G. (1997). No free lunch theorems for optimization. IEEE Transactions on Evolutionary Computation 1(1), 67–72.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Stefan Droste
    • 1
  • Thomas Jansen
    • 1
  • Ingo Wegener
    • 1
  1. 1.FB Informatik, LS 2Univ. DortmundDortmundGermany

Personalised recommendations