A Comparative Study of Adaptive Mutation Operators for Genetic Algorithms

Genetic algorithms (GAs) are a class of stochastic optimization methods inspired by the principles of natural evolution. Adaptation of strategy parameters and genetic operators has become an important and promising research area in GAs. Many researchers are applying adaptive techniques to guide the search of GAs toward optimum solutions. Mutation is a key component of GAs. It is a variation operator to create diversity for GAs. This paper investigates several adaptive mutation operators, including population level adaptive mutation operators and gene level adaptive mutation operators, for GAs and compares their performance based on a set of uni-modal and multi-modal benchmark problems. The experimental results show that the gene level adaptive mutation operators are usually more efficient than the population level adaptive mutation operators for GAs.