Solving a more complicated(not really) puzzle with QLearning Demo

Say your a car and want to get ice cream, and also you want to avoid zombies. Which action would be best to take?

1.) Mapping Agent(car) coordinates/situation into states

In this case there may too many states to cover, so we will just be adding states randomly or states that are important:

State that gets icecream

State that dies in zombies

2.) Adding random scenarios as inputs (even if its incomplete that's ok, as long as we achieve a goal)

Syntax: state action reward next-state


Results: (Note: Press compute policy button multiple times until it gets the optimal policy)

Per Action:

Now we know that our algorithm has limitations, its not a minimax algorithm where you get all the possible outcome, and doesn't quite fit in infinite states, we have to abstract our data to a finite state (we use our imaginations unfortunately..)

Another solution for this problem