On the Behavior of Evolutionary Global-Local Hybrids with Dynamic Fitness Functions

  • Authors:
  • Roger Eriksson;Björn Olsson

  • Affiliations:
  • -;-

  • Venue:
  • PPSN VII Proceedings of the 7th International Conference on Parallel Problem Solving from Nature
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper investigates the ability of evolutionary global-local hybrid algorithms to handle dynamic fitness functions. Using a model where fitness functions vary in ruggedness as well as in whether changes occur gradually or abruptly, we evaluate the performance of Baldwinian and Lamarckian hybrid strategies and find them capable of locating and tracking a moving global optimum.