I see, how do you get probabilities from Q-Values, Is it ok to use softmax on Q-Values?

Incorrect :) What you get using DQN is exactly the probabilities to select each action.
1
1
Shaked Zychlinski 🎗️
Piotr Niewiński
·Follow
Nov 27, 2020
--
I see, how do you get probabilities from Q-Values, Is it ok to use softmax on Q-Values?
--
--
Written by Piotr Niewiński29 Followers
·9 Following
No responses yet
Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams