Modified BP Network in Breaking Away from Flat Areas Effectively
一种有效脱离平坦区的改进BP网络

This paper proposes a new rule to adjust the weight value,which ensures enough weight adjusted range when the output of neuron is close to 0 or 1.The new rule solves a difficult problem for standard BP network to break away from the flat area when it has entered,here its training speed is very slow.A experiment which uses a BPnetwork with single hidden layer to approximate a nonlinear function is presented in the paper.It verifies the superiority of the new rule in the training speed,and draws a conclusion ...