
Jordan Terry
TalkRL: The Reinforcement Learning Podcast
00:00
How to Multiplay Your Orel?
In some games, all agents make their move at once and the it swaps over the environment. One way to handle this problem would be to have an api that says, ok, well, you can step for one agent at a time and stet for the next age, and so on. The alternative of everyone steps together will cable what happens if you're playing chess or paper scissors. This is not widely used for learning purposes but could help people learn more effectively.
Transcript
Play full episode