Adaption of Operator Probabilities in Genetic Programming

  • Authors:
  • Jens Niehaus;Wolfgang Banzhaf

  • Affiliations:
  • -;-

  • Venue:
  • EuroGP '01 Proceedings of the 4th European Conference on Genetic Programming
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this work we tried to reduce the number of free parameters within Genetic Programming without reducing the quality of the results. We developed three new methods to adapt the probabilities, different genetic operators are applied with. Using two problems from the areas of symbolic regression and classification we showed that the results in these cases were better than randomly chosen parameter sets and could compete with parameter sets chosen with empirical knowledge.