Tweeted By @hardmaru
I wrote a short tutorial explaining three different methods for training self-play agents (tournament-style genetic algorithms, self-play PPO, and self-play via cooperation) using Slime Volleyball Gym as an example environment.https://t.co/7rxqW0vLnb https://t.co/j0HWLQuWza pic.twitter.com/L7zVFjN2wx
— hardmaru (@hardmaru) September 24, 2020