Blue inside, orange outside. Nobody tells it the rule.

A new network starts with random connections.

So its first guesses are mostly wrong.

One dot goes in. The signal flows forward.

It measures how wrong it was.

The error flows back. Every connection is nudged.

Same dot again: a little less wrong.

Now repeat for every dot, thousands of times.

Each nudge is tiny. Together they find the rule.

Each neuron only knows one straight line.

Together, they draw the border.

Dots it has never seen? It gets them right.

It learned the rule from examples alone.