Basic types of neural networks

Basic types and learning algorithms for neural networks:

At present, there are dozens of kinds of neural network models, which can be classified into three categories according to network structure : Feedforward Network, feedback network and self-organizing network .

feedforward Neural network refers to the hierarchical arrangement of neurons, which are composed of input layer, middle layer and output layer respectively. Neurons in each layer only accept input from the previous layer, and the back layer has no signal feedback on the front layer. The input mode is propagated sequentially through each layer, and finally the output layer is output. This kind of network structure is usually suitable for prediction, pattern recognition and nonlinear function approximation , typical forward neural network based on gradient algorithm neural network such as BP network, optimal regularization method such as SVM, RBF neural Network and limit learning machine neural network.

Feedback Network , also known as the return network, the input signal determines the initial state of the feedback system, the system after a series of state transfer gradually converge to equilibrium state, therefore, stability is one of the most important indicators of feedback network, more typical is the perceptron network, Hopfield Neural Network, Hamming belief via network, wavelet neural network bidirectional Contact Storage Network (BAM), Boltzmann machine .

self-Organizing neural network is no teacher learning network, it simulates the human brain behavior, according to the past experience automatically adapt to unpredictable environmental changes, because there is no teacher signal, such networks usually use the principle of competition to network learning.

Basic types of neural networks