A Survey on Adaptation Strategies for Mutation and Crossover Rates of Differential Evolution Algorithm

Abstract

Differential Evolution (DE), the well-known optimization algorithm, is a tool under the roof of Evolutionary Algorithms (EAs) for solving non-linear and non-differential optimization problems. DE has many qualities in its hand, which are attributing to its popularity. DE also is known for its simplicity in solving the given problem with few control parameters: the population size (NP), the mutation rate (F) and the crossover rate (C_r). To avoid the difficulty involved in setting of suitable values for NP, F and C_r many parameter adaptation strategies are proposed in the literature. This paper is to present the working principle of the parameter adaptation strategies of F and C_r. The adaptation strategies are categorized based on the logic used by the authors, and clear insights about all the categories are presented.