Abstract:The high dimension of the learning error function for BP networks and the difficult computation complexity are incestigated. A simple modified conjugation-gradient decent algorithm (MPARTAN) is proposed based on improving the gradient BP algorithm. That the computation complexity of this algorithm is not higher than that of the BP momentum algorithm. Compared with FR conjugation algorithm, this algorithm has better stability and fast speed quality of convergence. It is also investigated that the convergence theorems for this algorithm and comparison of the computing results by two computing examples for the promoted three algorithms: BP momentum algorithm, FR conjugation-gradient algorithm and the novel MPARTAN algorithm.
梁久祯, 何新贵, 黄德双. 前馈神经网络的一种简单共轭梯度学习算法[J]. 北京航空航天大学学报, 2000, 26(5): 596-599.
LIANG Jiu-zhen, HE Xin-gui, HUANG De-shuang. Simple Conjugation-Gradient BP Algorithm for Feedforward Neural Networks. JOURNAL OF BEIJING UNIVERSITY OF AERONAUTICS AND A, 2000, 26(5): 596-599.
[1]umeihart D E, Hinton G E, Williams R J. Learning internal representations by error propagation[A].In:Rumelhart D E, McClelland J L, eds. Parallel Distributed Proceeding[C]. Cambridge MA:MIT Press, 1986. 318~362.
[2]umeihart D E, Hinton G E, Williams R J. Learning representations by backpropagating errors[J]. Nature, 1986,323(6088):533~536.
[3]echtNielsen R. Theory of the backpropagation neural network[A].In:Proceedings of the 1989 International Joint Conference on Neural Networks [C].NewYork:IEEE Press,1989.593~599.
[4]hah B V , Buehler R J, Kempthorne O. Some algorithms for minimizing a function of several variables[J]. J Soc Indust Appl Math, 1964, 12 (1) :74~92.
[5]曙光,郑崇勋,刘明远. 前馈神经网络中的反向传播算法及其改进:进展与展望[J]. 计算机科学,1996, 23(1):76~79.
[6]玖茜,魏权龄. 非线性规划及其理论[M].北京:中国人民大学出版社,1994. 226~228.
[7]uang Deshuang. An analysis of structure properties for feedforward neural networks[A].In: 1998 Int Conf on Neural Networks and Brain Proceedings[C]. Beijing: Publishing House of Electronics Industry, 1998. 463~466.