卷积神经网络是一种特殊的深层的神经网络模型，它的特殊性体现在两个方面，一方面它的神经元间的连接是非全连接的， 另一方面同一层中某些神经元之间的连接的权重是共享的（即相同的）。它的非全连接和权值共享的网络结构使之更类似于生物 神经网络，降低了网络模型的复杂度（对于很难学习的深层结构来说，这是非常重要的），减少了权值的数量。-Convolution neural network is a kind of special deep neural network model, its particularity manifests in two aspects, on one hand its neurons connection is not completely connected, on the other hand the same layer in some neurons The weights of the connections are shared (ie, the same). Its non-fully connected and weight-sharing network structure makes it more similar to the biological neural network, reducing the complexity of the network model (it is very important for the deep structure that is difficult to learn), reducing the weight quantity.