(Supervised, Unsupervised & RL)
The real difference between supervised, unsupervised and reinforcement learning — in about half an hour.
A free, visual course on the difference between supervised, unsupervised and reinforcement learning — the one thing that separates them, how each works, where the lines blur (self-supervised, RLHF), and where each shows up.
It's the map behind every ML project — spam filters (supervised), customer segments (unsupervised), game-playing and robot AI (reinforcement) — and ChatGPT quietly uses all three.
Show the model labeled examples — input and correct output — and it learns to predict the label.
No labels at all — the model hunts for patterns and groups in the data on its own.
An agent tries actions, gets rewards or penalties, and learns a strategy that maximizes reward.