Square Robots
An example of how Reinforcement Learning, specifically Q-Learning works. I implemented Q-Learning with tabular method.
Please like and subscribe if you enjoyed it! Leave a comment below for any feedback or questions, thanks!
Source
AI learns to play SNAKE using Reinforcement Learning
Interesting watch! Would you care to explain how it is possible for the agent to find the Apple if he can only see 1 square other than by trial and error? If he sees only 1 square ahead and not know the location of the Apple until it stumbles on it how it can learn to plan the best route?
Instructive, great video
Correction: at 0:35 I said the simple algorithm is not an AI but someone pointed out that it should be. I also believe that it should be considered as an AI agent. What I was trying to say is that it's not a learning agent.
I have one question. Is it easier to make hard-coded program that plays snake or self learning AI?
code?
I’ve got to know, what is that song called?
Q: have you any video teaching this q learning way of programming
Awesome video
question: did you make the game using pygame?
Have you published the code somewhere? What's the best result on what size of field (so we could calculate a coverage)?