Our next topic's a fun one: reinforcement learning. We can actually use this idea with an example of Pac-Man. We can actually create a little intelligent Pac-Man agent that can play the game Pac-Man really well on its own. You'll be surprised how simple the technique is for building up the smarts behind this intelligent Pac-Man. Let's take a look!

So, the idea behind reinforcement learning is that you have some sort of agent, in this case Pac-Man, that explores some sort of space, and in our example that space will be the maze that Pac-Man is in. As it goes, it learns the value of different state changes within different conditions.

For example, in the preceding image, the state of Pac-Man might be defined by the fact that it has a ghost to the South, and a wall to the West, and empty spaces to the North and East, and that might define the current state of Pac-Man. The state changes it can take would be to move in a given direction. I can then learn the value of going in a certain direction. So, for example, if I were to move North, nothing would really happen, there's no real reward associated with that. But, if I were to move South I would be destroyed by the ghost, and that would be a negative value.

As I go and explore the entire space, I can build up a set of all the possible states that Pac-Man can be in, and the values associated with moving in a given direction in each one of those states, and that's reinforcement learning. And as it explores the whole space, it refines these reward values for a given state, and it can then use those stored reward values to choose the best decision to make given a current set of conditions. In addition to Pac-Man, there's also a game called Cat & Mouse that is an example that's used commonly that we'll look at later.

The benefit of this technique is that once you've explored the entire set of possible states that your agent can be in, you can very quickly have a very good performance when you run different iterations of this. So, you know, you can basically make an intelligent Pac-Man by running reinforcement learning and letting it explore the values of different decisions it can make in different states and then storing that information, to very quickly make the right decision given a future state that it sees in an unknown set of conditions.