## Visitor's Comments

## Visitors' Comments on Tutorial: QLearning

We have 10 comments on this tutorial

Suggestion > POMDPTutorial? (date: 2010-06-13)

By Pavan

Dear Mr.Teknomo, First of all, I would like to thank you for the simple tutorial on q-learning. It has proved very helpful. I was wondering if you would considering putting up a tutorial on how to use q-learning algorithm in a partially observable MDP. It would be a very big help. Thank You and looking forward to your response.

Answer > POMDPTutorial? (date: 2010-06-14)

By Kardi

Dear Pavan, Thank you for your suggestion. I will put that topic in the list of suggested tutorials. I welcome anyone who wants to contribute to write this tutorial for Revoledu.

Thank you > wonderful introduction (date: 2010-12-16)

By wangshijun

thanks for this wonderful tutorial

Question > Q-learning by example (date: 2011-01-20)

By salmon labeeb

first, i\'ll like to commend you for the job you are doing. my question is how can one use Q-learning for a VPN bandwidth management/optimization because that\'s my project topic & i\'ve been working on it since but there seems to be no way out. thanks

Thank you > feedback (date: 2011-03-05)

By favas ali

It was a very helpful tutorial as i\'m doing a project based on the q-learning algoritm

Thank you > Very good job !! (date: 2011-11-01)

By Paolo Poletti

I was looking for an introductory reinforcment learning tutorial. I've a Engineering Laurea degree with a Robotic thesis but many of the found document was too complicated to start. Your tutorial is perfect !

Question > Is there a solution for the practice question? (date: 2012-10-17)

By nick

thanks

Question > Maze (date: 2013-01-10)

By mostafa darezereshki

I Was Working On Project That I Must Implement The Maze Algorithm By Q-Learning In A Graphical Enviroment.But I Didn't Have Any Idea About What Q-Learning Is!!!With Your Tutorial I got It.Now I Have A Point Of Start:) Have You Any Resource That can Help Me To Implement the Maze Problem With Q-Learning? Thanks And Regards:):*

Question > Help with understanding Convergence calculation (date: 2014-08-30)

By Michael Hussar

Hi, I just purchased the Q-Learning tutorial through paypal. I think you did a great job. I have a question/problem though regarding the implementation of the Q learning algorithm. I thought the paid version of the tutorial would go into more detail about aspects of the code. I am attempting to implement the ReinforcementLearning.m code in C#. My problem is determining how to implement the Convergence part of the code. An array of inf can not be represented in C# so I'm trying to figure out another way to do it but I'm not entirely sure I understand the concept behind how we determine if convergence has been reached. I do appreciate that convergence means that the algorithm has reached a point where no additional exploration will significantly improve the values held by the Q Matrix. But I'm not clear on how this is calculated from your code. Could you possibly explain in pseudo code or conceptually how we actually calculate Convergence for all the Q Matrix values. Maybe if you program in C#, you could tell me how to implement the convergence part of the Matlab code in C#? Thanks!

> (date: )

By