Application of improved multi expression programming algorithm in function finding problem
HU Zu-hui, XIA Shi-xiong, NIU Qiang (School of Computer Science and Technology, China University of Mining and Technology, Xuzhou 221116, China)
To improve the efficiency of multi expression programming (MEP) algorithm, the fitness function, crossover strategy and mutation strategy of the basic MEP algorithm is studied and optimized, and an improved algorithm is proposed. The normalized root mean square error (RMSE) is adopted in the improved MEP algorithm. Probability interval is used to choose crossover operators, and mutation probability is dynamically adjusted with evolutionary generation and fitness value in evolution process. At last, the improved MEP algorithm is applied to solve function finding problem, and efficiency of the algorithm is verified. The experimental results show that the improved MEP algorithm can find the target function more quickly than the basic MEP algorithm, which proves the efficiency of the algorithm is enhanced.