采用混合策略的改进基因表达式编程

发布时间：2018-01-23 02:25

本文关键词： 基因表达式编程(GEP) 镜像替换重启机制克隆选择　出处：《计算机科学与探索》2017年01期 　论文类型：期刊论文

【摘要】：基因表达式编程(gene expression programming,GEP)是一种新型的进化算法,在函数发现领域具有很好的应用。针对传统GEP存在的不足,提出了一种采用混合策略的改进基因表达式编程算法(improved gene expression programming algorithm used by hybrid strategy,HSI-GEP)。主要有两点改进:(1)采用镜像和重启机制对种群中的较差个体进行替换,以提高种群的质量和多样性;(2)在原有锦标赛选择之前引入克隆选择,以提高算法对优质解的开采能力。与权威文献中改进的GEP算法关于函数发现问题的大量对比实验表明,HSIGEP算法求解质量高,收敛速度快,具有明显的竞争力。
[Abstract]:Gene expression programming (GP) is a new evolutionary algorithm. It has a good application in the field of function discovery, aiming at the shortcomings of traditional GEP. An improved gene expression programming algorithm based on hybrid strategy is proposed. Improved gene expression programming algorithm used by hybrid. Strategy. In order to improve the quality and diversity of the population, there are two main improvements in HSI-GEP1: 1) the use of mirror image and restart mechanism to replace the poor individuals in the population, so as to improve the quality and diversity of the population. 2) Clonal selection is introduced before the original tournament selection to improve the mining ability of the algorithm to the high quality solution. A large number of comparative experiments on the function discovery problem of the improved GEP algorithm in authoritative literature show that the proposed algorithm can be used to solve the problem of function discovery. The HSIGEP algorithm has the advantages of high quality, fast convergence and obvious competitive power.
【作者单位】：西安建筑科技大学信息与控制工程学院;
【基金】：国家自然科学基金陕西省自然科学基金~~
【分类号】：Q811.4;TP301.6
【正文快照】： 1 引言基因表达式编程(gene expression programming,GEP)[1-2]由Ferreira在2001年提出,是进化算法的新成果。GEP继承了遗传算法(genetic algorithm,GA)与遗传编程(genetic programming,GP)的优点,具有GA编码、操作的简洁性和GP求解复杂问题强有力的空间搜索能力,成为函数发

【相似文献】