Chapter 8
Building a Basic Neural Network
IN THIS CHAPTER
Considering the basic architecture
Defining the problem
Understanding the solution process
Chapter 7 introduces neural networks using the simplest and most basic neural network of all: the perceptron. However, neural networks come in a number of forms, each of which has advantages. Fortunately, all the forms of neural networks follow a basic architecture and rely on certain strategies to accomplish what they need to do. If you learn how a basic neural network works, you can figure out how more complex architectures operate. The first part of this chapter discusses the basics of neural network functionality — that is, what you need to know to understand how a neural network performs useful work. It explains neural network functionality using a basic neural network that you can build from scratch using Python.
The second part of the chapter delves into some differences between neural networks. For example, you discover in Chapter 7 that individual neurons fire after reaching a particular threshold. An activation function determines when the input is sufficient for the neuron to fire, so knowing which activator functions are ...