Deep Deterministic Policy Gradients

[@sharvani6133]

Thank you for the video!

[@BananaLassi]

i feel like i can graduate after watching this video

[@leo.y.comprendo]

Thanks for the video! I have a question, what is the loss function used for the policy network?

[@daniamartinez4817]

Thank you so much!

[@brunomelicio2248]

Very good explanation. Thank you very much! Keep up the good work.

[@simmingi1]

for what? for you? what a selfish person 😅😅

[@amrahmed2009]

Very well explained. Thank you

[@maxschumacher12]

Excellent explanation!

[@ParagMantri]

This is very well explained.

[@overgeared]

excellent, thanks

[@Firestorm-tq7fy]

the video was not bad but sry, this has nothing todo with continuous action spaces. you simply described actor-critir RL and not continous action spaces...

[@wesnaw100]

It's a bit confusing because he doesn't go into detail on what the actor network is outputting, but it is indeed outputting continuous actions.

[@Firestorm-tq7fy]

@@wesnaw100 sry, but no. Continues action spaces are formatted as 2 outputs generating a distribution (variance and mean)

[@kkyars]

@@Firestorm-tq7fy yes, and that is a continuous distrbution