0
$\begingroup$

In his lecture 5 of the course "Reinforcement Learning", David Silver introduced GLIE Monte-Carlo Control.

enter image description here

I understand that we do policy evaluation for one step and then policy improvement. My question is how does the improved policy come into play in this GLIE algorithm?

Is Gt (return) based on the policy somehow? is that where the new policy comes in? Asked another way, how are policy evaluation and policy improvement connected in this image?

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ Hello. Welcome to AI SE! I just want to let you know that you case use latex on this site, so you may want to format your math symbols it ;) $\endgroup$
    – nbro
    Commented Jul 1, 2021 at 11:09

1 Answer 1

Reset to default
0
$\begingroup$

The policy is used in determining the next sequence of state-action pairs in the next episode. This means that the policy is determining indirectly the next Gt

Improve this answer
$\endgroup$

You must log in to answer this question.

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