For example,Бобцов

APPLICATION OF THE DIRECTED MUTATION TO CELLULAR AUTOMATA GENERATION PROCESS

Annotation

Cellular automata are widely used for the simulation of discrete systems. However, in most cases creation of controlling cellular automata is done manually, empirically or by exhaustive search. A number of papers describe methods for automatic generation of finite automata and cellular automata using genetic programming. However, relatively simple genetic operators are used in these issues not taking into account the current test patterns and the population state that makes strong impact on the performance and convergence of these methods. This paper deals with the classical mutation operator applied to the process of cellular automata generation and directed mutation operator, designed to eliminate the above shortcomings. Both described operators are used in the adaptive genetic algorithm. The operator of directed mutation performs the analysis of the current chromosome, test pattern, and offers an optimal variant of mutation on the basis of the information received. The main differences and advantages as compared with the standard mutation operator are described. Testing on several training examples is performed; data about the resulting performance for genetic algorithm is presented.

Keywords

Articles in current issue