前馈神经网络的一种简单共轭梯度学习算法

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

1.
北京航空航天大学计算机系
2. 北京系统工程研究所

基金项目: 国家自然科学基金资助项目(69705001)

详细信息

作者简介:
梁久祯(1968-),男,山东东平人,博士生,100083,北京.

中图分类号: TP 18
计量
- 文章访问数: 3347
- HTML全文浏览量: 184
- PDF下载量: 1306
- 被引次数: 0
出版历程
- 收稿日期: 1999-03-25
- 网络出版日期: 2000-05-31

Simple Conjugation-Gradient BP Algorithm for Feedforward Neural Networks

1.
Beijing University of Aeronautics and Astronautics, Dept. of Computer Science and Engineering
2. Beijing Institute of System Engineering

摘要

摘要: 针对前馈神经网络学习误差函数维数高、计算复杂度大的特点,对梯度下降BP算法加以改进从而构造出一种简单共轭梯度下降算法(MPARTAN算法).该算法计算复杂度不高于动量BP算法, 与FR共轭梯度法相比,该算法的稳定性好,又具有共轭梯度法的优点,收敛速度快.文中给出了该算法的收敛定理,并用2个实验例子比较了动量BP算法、FR共轭梯度法和MPARTAN算法的计算结果.
- 神经网络 /
- 收敛 /
- 共轭梯度法 /
- BP算法 /
- MPARTAN算法
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.
- neural networks /
- convergence /
- conjugate gradient method /
- BP algorithm /
- MPARTAN algorithm

HTML全文

参考文献(1)

[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 backpropagating errors[J]. Nature, 1986,323(6088):533～536. [3]echtNielsen 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.

施引文献

资源附件(0)

访问统计