Figure 1 shows the middleware infrastructure in relation to the first application to use it (shown in bold), and future applications and future back-end systems (shown in gray).
With the implicit function relation, BP (Back Propagation) neural network can easily realize the mapping between input data and output data.
通过找出其隐式函数关系,误差反向传播神经网络可以实现输入和输出间的任意映射。
3
This throw back to the "unitary" character of architecture, in which every part is in relation with the others and together they give a sense to the project.