GEP Book

  Home
  News
  Author
  Q&A
  Tutorials
  Downloads
  GEP Biblio
  Contacts

  Visit Gepsoft

 

C. FERREIRA In H. J. Caulfield, S.-H. Chen, H.-D. Cheng, R. Duro, V. Honavar, E. E. Kerre, M. Lu, M. G. Romay, T. K. Shih, D. Ventura, P. P. Wang, Y. Yang, eds., Proceedings of the 6th Joint Conference on Information Sciences, 4th International Workshop on Frontiers in Evolutionary Algorithms, pages 614-617, Research Triangle Park, North Carolina, USA, 2002.

Mutation, Transposition, and Recombination: An Analysis of the Evolutionary Dynamics

Introduction
 
Everybody agrees that, by and large, evolution relies on variation coupled with some kind of selection, and all evolutionary algorithms explore these fundamental assets. However, there is no agreement concerning the best way to create variation. The first evolutionary algorithms relied on mutation only and their recent developments claim that recombination has no general advantage over mutation (for a review see, e.g. [1]). In genetic algorithms (GAs), however, recombination is considered the more powerful of the two operators [5]. More recently, Spears [6] tries to conciliate both views, attributing to mutation and recombination equally important roles: the roles of disrupters (mutation) and constructors (recombination). To make matters worse, the classical genetic programming (GP) approach stresses the prominence of crossover [4], whereas developmental genetic programming (DGP) is more inclined towards mutation [2].

In this work I show that, in GEP, mutation is by far the single most important operator whereas recombination has a very limited power.

Home | Contents | Previous | Next