Episode Q0 increases exponentially

28 views (last 30 days)
DAMODARAN B.K
DAMODARAN B.K on 16 Feb 2021
Edited: DAMODARAN B.K on 17 Feb 2021
Can anyone explain why episode Q0 in RL increases exponentially after convergence of reward to a suboptimal policy?

Answers (1)

Emmanouil Tzorakoleftherakis
Hello,
Please take a look at this answer for some suggestions. Normalizing observations, rewards, and actions can also help avoid situations like these.
Hope this helps
  1 Comment
DAMODARAN B.K
DAMODARAN B.K on 17 Feb 2021
Edited: DAMODARAN B.K on 17 Feb 2021
is episode Q0, criticnetwork output or target value?

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!