An efficient conjugate gradient based learning algorithm for multiple optimal learning factors of multilayer perceptron neural network

摘要

In this paper, a second order learning algorithm based on Conjugate Gradient (CG) method for finding Multiple Optimal Learning Factors (MOLFs) of multilayer perceptron neural network is proposed in details. The experimental results on several benchmarks show that, compared with One Optimal Learning Factor algorithm with Optimal Output Weights (lOLF-OWO) and Levenberg-Marquardt learning algorithm (LM), our proposed CG based MOLF method with optimal output weights which is also called MOLFCG-OWO algorithm has not only significantly faster convergence rate than that of lOLF and even super to that of LM learning algorithm for some datasets with much less computational time, but also more generalization capability than lOLF-OWO. Thus, MOLFCG-OWO algorithm is suggested better choice for some practical applications.