Dynamic Parameter Encoding for Genetic Algorithms

  • Authors:
  • Nicol N. Schraudolph;Richard K. Belew

  • Affiliations:
  • Computer Science & Engineering Department, University of California, San Diego, La Jolla, CA 92093-0114. nici@cs.ucsd.edu;Computer Science & Engineering Department, University of California, San Diego, La Jolla, CA 92093-0114. rik@cs.ucsd.edu

  • Venue:
  • Machine Learning
  • Year:
  • 1992

Quantified Score

Hi-index 0.00

Visualization

Abstract

The common use of static binary place-value codes for real-valued parameters of the phenotype in Holland's genetic algorithm (GA) forces either the sacrifice of representational precision for efficiency of search or vice versa. Dynamic Parameter Encoding (DPE) is a mechanism that avoids this dilemma by using convergence statistics derived from the GA population to adaptively control the mapping from fixed-length binary genes to real values. DPE is shown to be empirically effective and amenable to analysis; we explore the problem of premature convergence in GAs through two convergence models.