What are stochastic bandits?

A stochastic bandit is like choosing between different candy machines that sometimes give you extra candies, but you don’t always know which one is the best.

Imagine you're in a room with several candy machines. Each time you pick a machine, it gives you some number of candies, but how many depends on luck. Some machines might usually give 5 candies, others only 2 or 3, and sometimes even more! You want to get as many candies as possible, but you don’t know which machine is the best at first.

So, you try different machines a few times, learning which ones are better. Then, you start picking mostly the best one, but still keep trying others just in case they might be even better. That’s how stochastic bandits work: it's like playing a game where you learn by trying and adjusting your choices over time.

Each machine is like a bandit, a sneaky, candy-loving thief who sometimes gives you more than others. The word "stochastic" means things happen with some randomness or chance, just like how some machines might give you extra candies out of the blue!

So you're not sure which machine to pick at first, but as you play more, you get smarter about your choices. It's like learning to be a candy master!