Nir Weingarten

Well, I think reinforcement learning is something that's very intuitive to understand because

It's very similar to how we as people go about the world.

And I'll just give a very crude example.

So imagine a baby is born and that baby crawls around the world and starts to discover it.

And they don't know anything about the world yet.

So they explore their environment and they try out different stuff.

And then the baby crawls around the floor and finds a piece of candy.

And so maybe it's on a plate because he doesn't pick the candy off the floor, but he finds a piece of candy and the baby tastes it and it's sweet and it's tasty.

And the baby has been rewarded by this experience.

And the next time they'll see a piece of candy, they'll try to eat it.

And the next day, say that baby roams about and then they found a slice of lemon.

And they see the lemon, they try it out and the lemon is sour and they don't like it.

It's not tasty.

So the next time the baby sees a lemon, they're not going to try it.

That is reinforcement learning.

And as very brutally.

And what do I mean by that?

I mean, you have some sort of autonomous agent.

In this case, it's a toddler.

You let them explore their environment, which is try out different foods.