Multilayer feedforward neural network

Z Wang, C Di Massimo, MT Tham, and AJ Morris. Procedure for determining the topology of multilayer feedforward neural networks. Neural Networks, 7(2) 291-300, 1994. 

Z Wang, MT Tham, and AJ Morris. Multilayer feedforward neural networks A canonical form approximation of nonlinearity. Int. J. Control, 56(3) 655-672, 1992. 

The most common neural network architecture is the multilayer feedforward neural network (often called multilayer perceptron, MLP). From the point of view of developing the neural network topology, the main elements for feedforward neural networks are the number of hidden layers, the number of neurons in each layer, the activation functions, and the training algorithm. 

Homik, K., Stinchcombe, M., and White, H., Multilayer feedforward networks are universal approximators, Neural Networks 2(5), 359-36 (1989). 

Among the most wide-spread neural networks are feedforward networks, namely multilayer perceptron (MLP). This network type has been proven to be universal function approximators [11], Another important feature of MLP is the ability to generalization. Therefore MLP can be powerful tool for design of intrusion detection systems. 

Figure 3 Multilayer neural network with feedforward and error backpropagation.

Smits, J. R. M., Meissen, W. J., Buydens, L. M. C., and Kateman, G. (1994) Using artificial neural networks for solving chemical problems. Part I. Multilayer feedforward networks. Chemom. Intell. Lab. Syst. 22, 165-189. 

Another multilayer network type is the multilayer feedforward network (see Neural Networks in Chemistry) that is usually trained by the back-propagation algorithm (Figure 2). The architecture of such a network can be quite variable, depending on the number of input units, the number of neurons, and the number of output neurons. In the case of training... 

---