Deep Learning from the Basics

2. Perceptrons

This chapter describes an algorithm called a perceptron. Invented by the US researcher Frank Rosenblatt in 1957, it is from this traditional algorithm that neural networks (i.e., deep learning) originated and is thus a necessary first step to the more advanced study of both. This chapter will describe a perceptron and use one to solve easy problems. Throughout this process, you will familiarize yourself with the mechanics of perceptrons.

What Is a Perceptron?

A perceptron receives multiple signals as inputs and outputs one signal. The "signal" here "flows" like an electric current or a river. In the same way that an electric current flows through a conductor and pushes electrons forward, the signal in a perceptron makes ...