摘要
分类问题在前向神经网络研究中占有重要位置.本文利用几何方法给出一个二进制神经网络K(≥2)分类问题的新学习算法.算法通过训练点的几何位置与类别分析,建立一个四层前向神经网络,实现网络输入向量分类.本文算法的优点在于:保证学习收敛且收敛速度快于BP算法及已有的其他一些前向网络学习算法;算法可以确定神经网络的结构且能实现精确的向量分类.另外,算法所建神经网络由线性阀值单元组成,神经元突触权值和阀值均为整数,特别适合于集成电路实现.
Binary to binary mapping for classification plays an important role in the researches on feed-forward-neural-network learning. In this paper, the geometrical method is employed to work out a new algorithm to train binary neural networks for classification. By analysis of every training vertex's geometrical location, the algorithm always produces a neural network of four layers for a certain classification problem. The advantages of this algorithm are: it runs with guaranteed convergence and goes to converge much more quickly than BP and some other algorithms; it can determine the structure of the neural networks by learning so that a precise classification is carried out.In addition, every neuron generated by the algorithm employs a hard-limit activation function with integer synaptic weights, which makes the actual implementation by VLSI technology more facilitated.
出处
《软件学报》
EI
CSCD
北大核心
1997年第8期622-629,共8页
Journal of Software
基金
国家自然科学基金
国家863高科技项目
山东省自然科学基金
关键词
神经网络
二进制
分类
几何学习算法
Neural networks, algorithm, convergence, training, geometry