Journal
Scientific and technical journal of information technologies, mechanics and optics
UDK
Issue:8 (53)
Download PDF0 Kbyte
The paper proposes a new method for representation of automaton as an individual in an evolutionary algorithm based on a decision trees. Application of this method greatly reduces amount of memory needed for running evolutionary algorithm. The method of genetic programming based on the proposed representation is created. Efficiency of this method is illustrated on the example of solving one of the variants of “Artificial ant” problem.