作者: Paul Albuquerque , Bastien Chopard , Christian Mazza , Marco Tomassini
DOI: 10.1007/978-3-540-46239-2_1
关键词:
摘要: In this paper we study the role of program representation on properties a type Genetic Programming (GP) algorithm. specific case, which believe to be generic standard GP, show that way individuals are coded is an essential concept impacts fitness landscape. We give evidence ruggedness landscape affects behavior algorithm and find that, below critical population, whose size representation-dependent, premature convergence occurs.