Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (29): 27-30.DOI: 10.3778/j.issn.1002-8331.2008.29.007

• 博士论坛 • Previous Articles     Next Articles

Experience-learning mechanism of ethnic group evolution algorithm

CHEN Hao1,CUI Du-wu1,CUI YIN-an1,2,TAO Yong-qin1,2   

  1. 1.School of Computer Science and Engineering,Xi’an University of Technology,Xi’an 710048,China
    2.School of Electronic and Information Engineering,Xi’an Jiaotong University,Xi’an 710049,China
  • Received:2008-05-15 Revised:2008-07-23 Online:2008-10-11 Published:2008-10-11
  • Contact: CHEN Hao

族群进化算法的经验学习机制

陈 皓1,崔杜武1,崔颖安1,2,陶永芹1,2   

  1. 1.西安理工大学 计算机学院,西安 710048
    2.西安交通大学 电子信息工程学院,西安 710049
  • 通讯作者: 陈 皓

Abstract: Based on the classifying process according to the coding similarity among typical individuals,a kind of population organization mechanism,ethnic group,has been created in chaotic population,and a novel evolution model,ethnic group evolution algorithm(EGEA),is developed.Based on ethnic group evolution mechanism,some typical individuals can be sifted out from population conveniently and a kind of experiential knowledge can be discovered in it.In this paper,an experience-learning mechanism is used to improve EGEA and the simulations show the experience -learning mechanism is feasible and valid,which improves the searching efficiency of EGEA greatly.

Key words: genetic algorithm, ethnic group evolution algorithm, ethnic group experience learning mechanism

摘要: 族群是依据个体编码特征的相似性对群体进行分类后形成的一种群体结构化组织,基于该机制形成了一种新的进化模型—族群进化算法(ethnic group evolution algorithm,EGEA)。族群机制可有效调控群体结构,协调算法的全局搜索和局部搜索时间,同时利用其所具有的分类能力也可方便地获取群体中的典型个体。设计了族群的经验学习机制来挖掘蕴含于群体中的进化经验知识,并利用这些知识来引导群体的搜索,提高EGEA的收敛速度。仿真实验表明族群的经验学习机制不仅是可行的而且是有效的,它显著提高了EGEA的搜索效率。

关键词: 遗传算法, 族群进化算法, 族群经验学习机制