Genetic Algorithm with Upgrading Operator

In recent years, Genetic Algorithm (GA) has grown rapidly and extensively used in various fields. GA is a search method based on the paradigm of natural selection and natural genetics. With their special characteristics: a coding of parameter set, searching from a population, appropriate measure of fitness, and probabilistic transition rules, GA can perform evolving a solution for several types of problems. Although the basic operators, crossover and mutation, work so well in the vast majority of genetic algorithm implementations. However, an important problem still remains in balancing between exploration and exploitation in genetic search. This problem concerns to a selective pressure and population diversity. Selective pressure can have a decisive effect on the outcome of an evolutionary search. Higher the pressure, more speed up the convergence but perhaps on a local optimum. Conversely, lower the pressure, slower the convergence but more variation of population. Anyway, population diversity provides the raw materials for adaptation. Hence, to let different species emerge and coexist inside the evolving population for a sufficient amount of time, the new computational operators for GA are proposed in this paper. Concentrating a chance to avoid complete loss of the characters on the worst chromosomes, the upgrading operator is presented. Performance of combination of the basic GA and the proposed operator is compared with the pure basic GA. Results on the optimization functions and the real applications in feature selection problem provide valuable evidence that the proposed GA perform more robust than the basic GA. Key-Words: Genetic Algorithm, Exploration, Exploitation, Selective Pressure, Population Diversity, Upgrading Operator