Hamming 神经网络从功能上来看是最小Hamming 距离分类器．利用它能够完成不完整输入信息与所存储模式的最小汉明距离分类．

　　 Hamming 网络是一个双层神经网络，第一层网（即匹配子网络）是用来计算输入模式与该网络已经学习过的各样本之间的匹配测度．第二层网（即竞争子网络）接收从匹配子网络送来的未知模式与已存各样本的匹配测度，然后经过多次迭代运算就可以求得与输入模式相匹配的样本．-Hamming neural network from the functional point of view is the smallest Hamming distance classifier. Use it to complete incomplete input information storage model with the smallest Hamming distance classifier. Hamming network is a double neural network, the first network layer (ie matching sub-network) is used to calculate the input model and the network has been studied by the match between the sample test degrees. The second layer network (that is, competitive-network) from the receiver-matching network sent an unknown model has been deposited with the samples matching measure, Then after many iterative computation can be obtained with the input mode matching sample.