Syntéza Boolean Symmetry Function pomocí jakýchkoli evolučních algoritmů - komparativní studie

Abstract

This contribution introduces analytical programming, a novel method that allows solving various problems from the symbolic regression domain. Symbolic regression was firstly proposed by J. R. Koza in his genetic programming and by C. Ryan for grammatical evolution. This contribution explains the main principles of analytic programming, and demonstrates its ability to synthesise suitable solutions, called programs. It is then compared with genetic programming and grammatical evolution. This comparative study is concerned with three Boolean k-symmetry problems from Koza's genetic programming domain, which are solved by means of analytical programming. Here, two evolutionary algorithms are used with analytical programming: differential evolution and self-organizing migrating algorithm.