中文说明：资源描述Elman网络是 J. L. Elman于1990年首先针对语音处理问题而提出来的, 它是一种典型的局部回归网络( global feed for ward l ocal recurrent)。Elman网络可以看作是一个具有局部记忆单元和局部反馈连接的前向神经网络。Elman网络具有与多层前向网络相似的多层结构。它的主要结构是前馈连接, 包括输入层、 隐含层、 输出层, 其连接权可以进行学习修正;反馈连接由一组“结构 ” 单元构成,用来记忆前一时刻的输出值, 其连接权值是固定的。在这种网络中, 除了普通的隐含层外, 还有一个特别的隐含层,称为关联层 (或联系单元层 ) ;该层从隐含层接收反馈信号, 每一个隐含层节点都有一个与之对应的关联层节点连接。关联层的作用是通过联接记忆将上一个时刻的隐层状态连同当前时刻的网络输入一起作为隐层的输入, 相当于状态反馈。隐层的传递函数仍为某种非线性函数, 一般为 Sigmoid函数, 输出层为线性函数, 关联层也为线性函数。
English Description：
Application backgroundElman network is a typical global L J. (feed for ward recurrent ocal), which is proposed by L. Elman in 1990. The Elman network can be regarded as a forward neural network with local memory and local feedback connections. Elman network has a multilayer structure similar to that of a multi layer forward network. Its main structure is a feed-forward connection, including input layer, hidden layer and output layer, the connection weights can be corrected by a group of "structure" unit, which is used to remember the output value of the previous one. In this kind of network, in addition to the ordinary hidden layer, there is a special hidden layer, called the association layer (or contact element layer); the layer from the hidden layer to receive feedback signals, each hidden layer nodes have a corresponding layer node connection. The function of the correlation layer is to be used as an input of the hidden layer, which is equivalent to the state f