From Perceptrons to Neural Networks
A neural network is similar to the perceptron described in the previous chapter in many ways. How a neural network works, as well as how it differs from a perceptron, will be described in this section.
Neural Network Example
Figure 3.1 shows a neural network example. Here, the left column is called an input layer, the right column is called an output layer, and the center column is called the middle layer. The middle layer is also known as a hidden layer. "Hidden" means that the neurons in the hidden layer are invisible (unlike those in the input and output layers). In this book, we'll call the layers layer 0, layer 1, and layer 2 from the input layer to the output layer (layer numbers start from layer 0 because doing so is convenient when the layers are implemented in Python later). In Figure 3.1, layer 0 is the input layer, layer 1 is the middle layer, and layer 2 is the output layer:
