Asymmetric mutation model in genetic algorithm

Genetic algorithms (GAs) are stochastic optimization techniques, and the theoretical study of the process of GA evolution is very important in the application of GA. Mutation is one of most important operators in GA, and Markov chain theory has attracted researchers’ attention for the study of mutation. By applying Markov chain to study symmetric mutation model in GA, we have obtained the relation between the mutation rate and the evolution of the first order schema. This paper theoretically analyzes the effects of mutation rates on GA with asymmetric mutation, and studies the evolution and stationary distribution of the first order schema. This study focuses on effects of asymmetry to the linkage of loci, and shows the degree of asymmetry in mutation has a large effect on the evolution of the first order schema.