首页    期刊浏览 2024年11月26日 星期二
登录注册

文章基本信息

  • 标题:多層個体群を有する遺伝的プログラミングの提案およびBoolean問題への適用
  • 本地全文:下载
  • 作者:長谷川 拓 ; 森 直樹 ; 松本 啓之亮
  • 期刊名称:進化計算学会論文誌
  • 电子版ISSN:2185-7385
  • 出版年度:2017
  • 卷号:8
  • 期号:2
  • 页码:52-60
  • DOI:10.11394/tjpnsec.8.52
  • 语种:Japanese
  • 出版社:The Japanese Society for Evolutionary Computation
  • 摘要:

    In Evolutionary Computation (EC), it is difficult to maintain efficient building blocks and to combine them efficiently. In particular, the control of building blocks in the population of Genetic Programming (GP) is relatively difficult because of tree-shaped individuals and also because of bloat, which is the uncontrolled growth of ineffective code segments in GP. It has been reported that the parameter tuning for solving the above mentioned problems requires a significant amount of efforts. Aimed at utilizing building blocks efficiently, this paper presents a GP algorithm called “Genetic Programming with Multi-Layered Population Structure (MLPS-GP)”. MLPS-GP employs multi-layered population like the pyramid-like population and searches solutions using local search and crossover. The computational experiments were conducted by taking several classical Boolean problems as examples.

  • 关键词:genetic programming;population structure
国家哲学社会科学文献中心版权所有