分布估计算法综述  

A Survey on Estimation of Distribution Algorithms

作　　者：周树德[1] 孙增圻[1] 

机构地区：[1]清华大学计算机系智能技术与系统国家重点实验室,北京100084

出　　处：《自动化学报》2007年第2期113-124,共12页Acta Automatica Sinica

基　　金：国家重点基础研究专项基金(G2002cb312205)资助~~

摘　　要：分布估计算法是进化计算领域新兴起的一类随机优化算法,是当前国际进化计算领域的研究热点．分布估计算法是遗传算法和统计学习的结合,通过统计学习的手段建立解空间内个体分布的概率模型,然后对概率模型随机采样产生新的群体,如此反复进行,实现群体的进化．分布估计算法中没有传统的交叉、变异等遗传操作,是一种全新的进化模式;这种优化技术能够通过概率图模型对变量之间的关系进行建模,从而能有效的解决多变量相关的优化问题．根据概率模型的复杂性,本文按照变量无关、双变量相关、多变量相关等三类分别介绍相应的分布估计算法．作为一篇综述性文章,本文旨在全面系统的向国内读者介绍这一新技术,并总结分布估计算法的研究现状和未来的研究方向．Estimation of distribution algorithms （EDAs） are a class of novel stochastic optimization algorithms, which have recently become a hot topic in field of evolutionary computation. EDAs acquire solutions by statistically learning and sampling the probability distribution of the best individuals of the population at each iteration of the algorithm. EDAs have introduced a new paradigm of evolutionary computation without using conventional evolutionary operators such as crossover and mutation. In such a way, the relationships between the variables involved in the problem domain are explicitly and effectively exploited. According to the complexity of probability models for learning the interdependencies between the variables from the selected individuals, this paper gives a review of EDAs in the order of interactions： dependency-free, bivariate dependencies, and multivariate dependencies, aiming to bring the reader into this novel flied of optimization technology. In addition, the future research directions are discussed.

关 键 词：分布估计算法 遗传算法 统计学习 概率模型 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]