Take the 2-minute tour ×
Computer Science Stack Exchange is a question and answer site for students, researchers and practitioners of computer science. It's 100% free, no registration required.

I am new to the reinforcement learning. In the textbook there is a question which confuses me. It says: To think about three substantially different reward schemes for a given policy on the grid world, and for each scheme I should give rewards and discount.

I can think of one, but it needs three substantially different reward schemes. What could others look like?

Picture:

enter image description here

share|improve this question

migrated from cstheory.stackexchange.com Aug 25 '12 at 22:42

This question came from our site for theoretical computer scientists and researchers in related fields.

    
Please consider adding the picture directly (i.e. uploading it to imgur), thus placing it under a Creative Commons license. –  Raphael Sep 25 '12 at 6:00
1  
Please, give more information - your question is very unclear now. With a simple search I found another question of yours about this grid world with much more details, but yet not enough for this question. –  Anton Nov 24 '12 at 7:52
    
you need to specify what "S", "G" is in your picture and what the arrows represent. –  Strin Dec 24 '12 at 10:23

1 Answer 1

The reward scheme determines what your agent is to be learning. There is no limit.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.