【摘要】 基因表达式程序设计(Gene Expression Programming,简称GEP)与遗传程序设计(Genetic Programming,简称GP)相比,具有更强的搜索能力、更简单的编码表示方法和产生更高复杂性函数的能力.但是它也存在一些缺点,例如缺乏学习机制,搜索过于盲目.针对这一缺点,本文提出了一种新的自动程序设计方法:基因评估基因表达式程序设计(Gene Estimated Gene Expression Programming,简称GEGEP).与GEP相比,GEGEP主要具有如下特点:(1)改变了GEP的基因表达式结构,将原来的“头+尾”结构改成了“头+身+尾”结构,以利于其引进学习机制;(2)同源基因也采用“头+身+尾”结构,以利于增强其搜索能力;(3)引入了分布评估算法(Estimation of Distribution Algorithm,简称EDA)的思想,以利于增加其学习能力并且加快其收敛速度.实验结果表明,与GEP和GP相比,GEGEP具有更高的拟合和预测精度、更快的收敛速度.更多还原

【Abstract】 Gene Expression Programming (GEP) has wider searching ability,simpler representation and the creation of higher levels of complexity than GP. However, it has some shortcomings, such as blind searching without learning mechanism. So we propose a new automatic programming method: Gene Estimated Gene Expression Programming (GEGEP). Compared with GEP, it mainly has the following characteristics: First, improve the gene expression structure, original gene structure is divided the head of chromosomes into a head and a body, which can be used to introduce the learning mechanism. Second, the homeotic gene which is also composed of a head, a body and a tail is used, which can increase its searching ability. Third, the idea of EDA is introduced, which can enhance its learning ability and accelerate the convergence rate. The results of experiments show that GEGEP has better fitting and predicted precision, faster convergence speed than GP and GEP.更多还原