1
$\begingroup$

I would really like to see an example of Q-learning that I could read, so that I can learn Q-learning from scratch. I read some articles on the internet, but I found it a little difficult to understand step by step, as most articles explain it using libraries.

But what I would really like is to be able to understand the basics of how Q-learning works from scratch.

Please, could someone give a very simple practical example of Q-learning, in a very small environment (like a 3 x 3 grid)?, a very simple step-by-step that could be done manually?

$\endgroup$
1
  • 5
    $\begingroup$ The trouble with this question, and why you may not get any answers, is that it is not clear what is blocking you from learning. Someone who goes to a lot of trouble to write you a tutorial may also have to spend hours guiding you through it and re-writing it. Have you looked at this book: incompleteideas.net/book/the-book-2nd.html - it's a free to download PDF, and contains very good introduction to reinforcement learning. It breaks things down in great detail, and Q-learning isn't until page 133 . . . I suggest you read that from the beginning and ask questions when you are stuck $\endgroup$ Nov 4, 2023 at 17:48

1 Answer 1

2
$\begingroup$

one great library is called openAI gym which has a bunch of toy problems. One of these is CliffWalking based on Sutton and is a 3x12. This article steps through it pretty nicely and I am sure you can also find other public github repos that solve it. Honestly, you should do it by hand with easy numbers like 100 as the reward and reduce it by .5 every step away or somthing similar.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged .