ML Anthology

Dynamic Parameter Encoding for Genetic Algorithms

Nicol N. Schraudolph, Richard K. Belew
MLJ 1992 pp. 9-21
doi:10.1007/BF00993252 /mlj/1992/schraudolph1992mlj-dynamic/

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.

PDF MLJ Semantic Scholar

Cite

Text

Schraudolph and Belew. "Dynamic Parameter Encoding for Genetic Algorithms." Machine Learning, 1992. doi:10.1007/BF00993252

Markdown

[Schraudolph and Belew. "Dynamic Parameter Encoding for Genetic Algorithms." Machine Learning, 1992.](https://mlanthology.org/mlj/1992/schraudolph1992mlj-dynamic/) doi:10.1007/BF00993252

BibTeX

@article{schraudolph1992mlj-dynamic,
  title     = {{Dynamic Parameter Encoding for Genetic Algorithms}},
  author    = {Schraudolph, Nicol N. and Belew, Richard K.},
  journal   = {Machine Learning},
  year      = {1992},
  pages     = {9-21},
  doi       = {10.1007/BF00993252},
  volume    = {9},
  url       = {https://mlanthology.org/mlj/1992/schraudolph1992mlj-dynamic/}
}