AI School › How it’s built inside · Lesson 62 of 130

🥞 The layers of the network

Neurons are stacked in layers, like a tiered cake.

Artificial neurons are not thrown together randomly: they are organized in layers, one above another, like the tiers of a cake or the steps of an assembly line. Information enters at the first layer, moves to the second, then the third, and so on to the last. At each level something more refined happens. Picture a bicycle factory: the first station assembles the wheels, the second adds the frame, the third the handlebars, and at the last the bike is ready. No station does everything alone, but together they build something complex.

In a network that recognizes images it works exactly like this. The first layer notices only simple things, like little lines and color patches. The second combines those lines and spots shapes, like a circle or a corner. The third understands that those circles and corners form an eye, an ear. As you climb, the network moves from tiny details to bigger and bigger ideas, until it says "this is a cat." The more layers there are, the more complicated things the network can grasp: that is why we call it "deep" learning, meaning many stacked levels.

🎒 Try it in class (or at home)

Draw a three-station assembly line that builds a snowman: what does each station do?

← Lesson 61: Artificial neurons · Lesson 63: What tokens are →

From our network

Magellano GPS: Fleet Tracking Made Simple

Real-time GPS tracking, remote engine lock, fuel and CO₂ reporting for your fleet.

Visit magellanogps.com →

From the Agora Intelligence blog