AlphaStar: Grandmaster level in StarCraft II using multi-agent reinforcement learning

Any topic that does not fit elsewhere
Post Reply
Rémi Coulom
Posts: 120
Joined: Tue Feb 12, 2008 8:31 pm
Contact:

AlphaStar: Grandmaster level in StarCraft II using multi-agent reinforcement learning

Post by Rémi Coulom » Thu Oct 31, 2019 12:19 pm

https://www.deepmind.com/blog/article/A ... t-learning

Direct link to the paper: https://storage.googleapis.com/deepmind ... matted.pdf

The most interesting contribution seems to be their league system to reach the Nash equilibrium. Their method for off-policy learning is very interesting, too. This gives me some inspiration for my mahjong reinforcement-learning experiments.

Post Reply