The union optimization method of multilayer perceptron

Abstract

The backpropagation algorithm (BP) is one of the most popular algorithms in neural network studies, but it is notorious for its slowness in achieving convergence. Based on the theory of unconstrained optimization, the union optimization method (UOM) is presented. The method consists of two parts: apply Descent Method to optimize objective function E=/spl Sigmasub p/E/sub p/; when E>