Lozi map generated initial population in analytical programming

Abstract

Analytical programming is a novel approach to symbolic regression independent on the used evolutionary algorithm. This research paper focuses on the usage of Lozi chaotic map based pseudo-random number generator for the generation of the initial population of the selected evolutionary algorithm. The researched benefit is the tendency to generate individuals which are mapped to more complex programs than that of individuals generated by classical pseudo-random number generator. The results show that there is a potential in replacing classical generator by the chaotic map based one in order to generate more complex programs.