5 comments

[ 1.7 ms ] story [ 24.3 ms ] thread
just in time for Q*
Maybe I'm pointing out the obvious, but there are references to Q* in the docs[1]. Specifically it's just the Bellman equation describing the optimal action-value function.

It would be kind of weird if OpenAI invented their own thing called Q* which collides with an existing name in a topic I'd expect them to be familiar with.

https://spinningup.openai.com/en/latest/algorithms/ddpg.html

It's a cool resource but all the examples have bitrotted to the point that I gave up trying to get them to work.