Estimation Algorithm Based On Evolutionary Programming Distribution

Evolutionary Programming is a kind of stochastic optimization algorithm. EP algorithms are based on an arbitrarily initialized population of search points which evolves towards better and better regions in the search space by means of randomized process of mutation and selection. Premature convergence is also a big problem of evolutionary programming. So avoiding premature convergence and balancing the ability of exploration and exploitation has become one of the important aspects of EP's study. Estimation of distribution algorithm is a new class of evolutionary algorithms, which combines the statistic theory with evolutionary schemes.The major works in this paper are: This paper presents a brief overview of Evolutionary Programming and Estimation of distribution algorithm, and clears the main ideas, planning, convergence and the advantages and disadvantages of Evolutionary Programming and Estimation of distribution algorithm. Based on this, advances a new algorithm, Estimation of distribution algorithms based Evolutionary Programming, EDA can use the EP in the evolutionary process of populations were estimated probability distribution to predict the optimal individuals, provide the search direction for the EP, accelerate the convergence. Then its convergence is proved.Finally we use the new algorithm to solve optimization problems, and eight benchmarks are chosen in our experiments.