Summary
In this chapter, we covered perceptrons. The perceptron is a very simple algorithm, so you should be able to understand how it works quickly. The perceptron is the basis of a neural network, which we will learn about in the next chapter. These points may be summed up in the following list:
- A perceptron is an algorithm with inputs and outputs. When it receives a certain input, it outputs a fixed value.
- A perceptron has "weight" and "bias" parameters.
- You can use perceptrons to represent logic circuits such as AND and OR gates.
- An XOR gate cannot be represented with a single-layer perceptron.
- A two-layer perceptron can be used to represent an XOR gate.
- A single-layer perceptron can only represent linear areas, while a multilayer perceptron can represent nonlinear areas.
- Multilayer perceptrons can represent a computer (theoretically).
