386771 (3) [Avatar] Offline
#1
Great book so far.

I'm having problems to understand this equation in the box:

Ra(r) = P[R=r | A=a]

Shouldn't be something like "R(a)"?

Thanks
Miguel Morales (15) [Avatar] Offline
#2
Good feedback, I think this equation can be simplified.

R^a(r) is a probability (say 0.34), of the reward r (say +5.353) occurring when taking action 'a' (say action 2).

There is definitely room for improvement/clear explanation. Will make a note. Thanks for the feedback.