The importance of mutation varies across evolutionary computation domains including: genetic programming, evolution strategies, and genetic algorithms. In the genetic programming community, researchers' view of mutation's effectiveness spans the range from an ineffective or marginal operator, to a neutral operator, to a highly effective operator that evolves solutions more effectively than genetic programming with crossover alone. Mutation implementation and associated parameters are often under reported in genetic programming research and typically lack context that justifies the technique and parameter selection. In part, reporting variance stems from the adaptation of mutation developed by the genetic algorithm community, and the creation of new mutation techniques in genetic programming. This survey describes the controversial operator in genetic programming applications, mutation selection operators, mutation techniques and offers an organization of mutation characteristics. We suggest methodologies to improve reporting of mutation parameters and related individual selection methods.
展开▼
