Fast evolutionary programming through search momentum and multipleoffspring strategy

摘要

A new algorithm that helps to accelerate convergence as well as toenhance the diversity of the evolutionary programming (EP) searchtechnique is proposed, based on an individual structure concept. Themajor components of the algorithm that lie behind its good performanceincludes scaling, selection strategy, the use of age and the searchdirection (or momentum) vector, and multiple offspring per parent. Notonly are the multiple offspring approach and the search direction vectorconcept novel but the combination of these features used for EP is alsonew. Through a benchmark test, its search performance has been found tobe superior to the conventional EP and one of its acceleration methods