Skip to main content
João Pedro's user avatar
João Pedro's user avatar
João Pedro's user avatar
João Pedro
  • Member for 4 years, 8 months
  • Last seen more than a month ago
Loading…
Loading…
comment
Derivation of Monte Carlo Policy Gradient for REINFORCE
Thanks for the answer. In fact 1, your proof is for a conditional expected value giving a particular value $Y=y$, but in PG formula the it's conditioned on RV $Y$. I am too struggling with this derivation on the book. RL course on YT based on the book usually skip this derivation which is kind of sad...
awarded
awarded
awarded
Loading…
Loading…