计算机应用 ›› 2010, Vol. 30 ›› Issue (07): 1896-1898.

• 人工智能 • 上一篇    下一篇

新的线性遗传程序设计方法

潘小海1,徐蔚鸿2,周恺卿3   

  1. 1. 长沙理工大学计算机与通信工程学院
    2. 长沙理工大学 计算机与通信工程学院
    3. 长沙理工大学
  • 收稿日期:2010-01-04 修回日期:2010-03-08 发布日期:2010-07-01 出版日期:2010-07-01
  • 通讯作者: 潘小海
  • 基金资助:
    模糊神经网络簇的通用有效学习算法及其新型鲁棒性的研究;模糊神经网络簇有效学习算法及其鲁棒性研究

New linear genetic programming approach

  • Received:2010-01-04 Revised:2010-03-08 Online:2010-07-01 Published:2010-07-01

摘要: 受其他多种线性编码的遗传程序设计算法的启发,提出一种新的编码方式的遗传程序设计——符号遗传程序设计。该编码方式具有简单、无语法限制并且能够在不增加计算量的情况下将染色体翻译成多个表达式等特点。分析与实验表明该算法具有较高的效率和较强的稳定性。

关键词: 遗传程序设计, 进化算法, 符号回归, 线性编码

Abstract: A new genetic programming named Symbol Genetic Programming (SGP) based on a new encoding method was proposed. This new encoding method absorbed the merits of many other linear genetic programming methods. It coded with a simple, unrestrained string. Based on its characteristic, multi-expressions could be contained in one individual without the increase of computation task. This method is proved to be effective and stable through the complexity analysis and experiment.

Key words: Genetic Programming, Linear Representation, Evolutionary Algorithm, Symbolic Regression