+ Create new tricision Login 

Login

or


Forgot password?

First visit? Sign-up now!

By continuing you accept the Terms & Conditions and Privacy Policy.

time is up

Question about Prediction with Approximation?

saved
Ideas
Pros and cons
 
Votes
I'm finding sections 9.2 - 9.5 very challenging to grasp, even after watching the videos and reading most of the chapter. Can we spend a bit more time on it this week?
 
6

Adam V, Adam IV, Not Adam and 3 more

Why is Reward = +1 / step?
 
Because the value function here means: How many steps to the goal?
by Martha
0
Why is there no mu(s) in the gradient term? Is that also absorbed in alpha?
0
Are there methods to learn a good state space representation, and what states should be combined? Maybe offline using data collected in the past?
0

Comments

http://www.tricider.com/brainstorming/3D4V06mUv2V