Advertisement

Performance Measures for Dynamic Environments

  • Karsten Weicker
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2439)

Abstract

This article investigates systematically the utility of performance measures in non-stationary environments. Three characteristics for describing the goals of a dynamic adaptation process are proposed: accuracy, stability, and recovery. This examination underpins the usage of the best fitness value as a basis for measuring the three characteristics in scenarios with moderate changes of the best fitness value. However, for dynamic problems without coordinate transformations all considered fitness based measures exhibit severe problems. In case of the recovery, a newly proposed window based performance measure is shown to be best as long as the accuracy level of the optimization is rather high.

Keywords

Genetic Algorithm Dynamic Environment Dynamic Problem Dynamic Optimization Parallel Problem 
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. Angeline, P. J. (1997). Tracking extrema in dynamic environments. In P. J. Angeline, R. G. Reynolds, J. R. McDonnell, & R. Eberhart (Eds.), Evolutionary Programming VI (pp. 335–345). Berlin: Springer. (Lecture Notes in Computer Science 1213)CrossRefGoogle Scholar
  2. Branke, J. (1999). Evolutionary algorithms for dynamic optimization problems: A survey (Tech. Rep. No. 387). Karlsruhe, Germany: Institute AIFB, University of Karlsruhe.Google Scholar
  3. Cobb, H. G. (1990). An investigation into the use of hypermutation as an adaptive operator in genetic algorithms having continuous, time-dependent nonstationary environments (Tech. Rep. No. 6760 (NLR Memorandum)). Washington, D.C.: Navy Center for Applied Research in Artificial Intelligence.Google Scholar
  4. Collard, P., Escazut, C., & Gaspar, A. (1997). An evolutionary approach for time dependant optimization. International Journal on Artificial Intelligence Tools, 6 (4), 665–695.CrossRefGoogle Scholar
  5. Dasgupta, D., & McGregor, D. R. (1992). Nonstationary function optimization using the structured genetic algorithm. In R. Männer & B. Manderick (Eds.), Parallel Problem Solving from Nature 2 (Proc. 2nd Int. Conf. on Parallel Problem Solving from Nature, Brussels 1992) (pp. 145–154). Amsterdam: Elsevier.Google Scholar
  6. De Jong, K. (2000). Evolving in a changing world. In Z. Ras & A. Skowron (Eds.), Foundation of intelligent systems (pp. 513–519). Berlin: Springer.Google Scholar
  7. De Jong, K. A. (1975). An analysis of the behavior of a class of genetic adaptive systems. Unpublished doctoral dissertation, University of Michigan.Google Scholar
  8. Feng, W., Brune, T., Chan, L., Chowdhury, M., Kuek, C. K., & Li, Y. (1997). Benchmarks for testing evolutionary algorithms (Tech. Rep. No. CSC-97006). Glasgow, UK: Center for System and Control, University of Glasgow.Google Scholar
  9. Goldberg, D. E., & Smith, R. E. (1987). Nonstationary function optimization using genetic algorithms with dominance and diploidy. In J. J. Grefenstette (Ed.), Proc. of the Second Int. Conf. on Genetic Algorithms (pp. 59–68). Hillsdale, NJ: Lawrence Erlbaum Associates.Google Scholar
  10. Grefenstette, J. J. (1992). Genetic algorithms for changing environments. In R. M:anner & B. Manderick (Eds.), Parallel Problem Solving from Nature 2 (Proc. 2nd Int. Conf. on Parallel Problem Solving from Nature, Brussels 1992) (pp. 137–144). Amsterdam: Elsevier.Google Scholar
  11. Grefenstette, J. J. (1999). Evolvability in dynamic fitness landscapes: a genetic algorithm approach. In 1999 Congress on Evolutionary Computation (pp. 2031–2038). Piscataway, NJ: IEEE Service Center.Google Scholar
  12. Hadad, B. S., & Eick, C. F. (1997). Supporting polyploidy in genetic algorithms using dominance vectors. In P. J. Angeline, R. G. Reynolds, J. R. McDonnell, & R. Eberhart (Eds.), Evolutionary Programming VI(pp. 223–234). Berlin: Springer. (Lecture Notes in Computer Science 1213)CrossRefGoogle Scholar
  13. Lewis, J., Hart, E., & Ritchie, G. (1998). A comparison of dominance mechanisms and simple mutation on non-stationary problems. In A. E. Eiben, T. Bäck, M. Schoenauer, & H.-P. Schwefel (Eds.), Parallel Problem Solving from Natur—PPSN V (pp. 139–148). Berlin: Springer. (Lecture Notes in Computer Science 1498)CrossRefGoogle Scholar
  14. Mori, N., Kita, H., & Nishikawa, Y. (1996). Adaptation to a changing environment by means of the thermodynamical genetic algorithm. In H. Voigt, W. Ebeling, & I. Rechenberg (Eds.), Parallel Problem Solving from Nature— PPSN IV (Berlin, 1996) (Lecture Notes in Computer Science 1141) (pp. 513–522). Berlin: Springer.CrossRefGoogle Scholar
  15. Mori, N., Kita, H., & Nishikawa, Y. (1998). Adaptation to a changing environment by means of the feedback thormodynamical geentic algorithm. In A. E. Eiben, T. Bäck, M. Schoenauer, & H.-P. Schwefel (Eds.), Parallel Problem Solving from Nature— PPSN V (pp. 149–158). Berlin: Springer. (Lecture Notes in Computer Science 1498)CrossRefGoogle Scholar
  16. Salomon, R., & Eggenberger, P. (1997). Adaptation on the evolutionary time scale: A working hypothesis and basic experiments. In J.-K. Hao, E. Lutton, E. Ronald, M. Schoenauer, & D. Snyders (Eds.), Artificial Evolution: Third European Conf., AE’97 (pp. 251–262). Berlin: Springer.Google Scholar
  17. Trojanowski, K., & Michalewicz, Z. (1999). Searching for optima in non-stationary environments. In 1999 Congress on Evolutionary Computation (pp. 1843–1850). Piscataway, NJ: IEEE Service Center.Google Scholar
  18. Vavak, F., Fogarty, T. C., & Jukes, K. (1996). A genetic algorithm with variable range of local search for adaptive control of the dynamic systems. In Proc. of the 2nd Int. Mendelian Conf. on Genetic Algorithms.? Google Scholar
  19. Weicker, K. (2000). An analysis of dynamic severity and population size. In M. Schoenauer, K. Deb, G. Rudolph, X. Yao, E. Lutton, J. J. Merelo, & H.-P. Schwefel (Eds.), Parallel problem solving from nature— PPSN VI (pp. 159–168). Berlin: Springer.CrossRefGoogle Scholar
  20. Weicker, K., & Weicker, N. (1999). On evolution strategy optimization in dynamic environments. In 1999 Congress on Evolutionary Computation (pp. 2039–2046). Piscataway, NJ: IEEE Service Center.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Karsten Weicker
    • 1
  1. 1.Institute of Computer ScienceUniversity of StuttgartStuttgartGermany

Personalised recommendations