Introduction and Single-Layer Neural Networks

Neural networks are potentially massively parallel distributed structures and have the ability to learn and generalize. The neuron is the information processing unit of a neural network and the basis for designing numerous neural networks. The most fundamental network architecture is a single-layer neural network, where the ?>single-layer?> refers to the output layer of computation neurons. This chapter introduces Rosenblatt's neuron. Rosenblatt's perceptron occupies a special place in the historical development of neural networks. The chapter also considers the performance of the perceptron network and is in a position to introduce the perceptron learning rule. This learning rule is an example of supervised training, in which the learning rule is provided with a set of examples of proper network behavior. Finally the chapter further discusses activation function and its types, including a threshold function, or Heaviside function and sigmoid function.