-
Notifications
You must be signed in to change notification settings - Fork 456
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Exercise4.9 #81
Comments
I got the same result as @soonjune . |
Does anybody know why this happens, I get similar results with this implementation and my own. Is this because it does not result in a stable solution? |
Due to the limited accuracy, it is not reliable to compare equality, some value is actually the same. So we should limit the precision when we compare the value of q(s,a)——> just change the part of "np.argmax" |
I think there is something wrong with the implementation. The optimal policy should be 1 for all states when p_h = 0.55. There is a big bet around 80 and I cannot find any reason for this behavior.
The text was updated successfully, but these errors were encountered: