League play in AlphaStar — mechanism.
hardAnswer
- Vinyals et al. 2019.
- Population of agents in 3 roles: (1) Main agents — general strong players.
- (2) Main Exploiters — target Main agents specifically.
- (3) League Exploiters — cover blind spots across whole league.
- All train against a mixture of past + current opponents.
- Prevents strategy cycling (rock-paper-scissors problem).
- Grandmaster StarCraft II.
- Foundation of modern multi-agent training for esports.
Check yourself — multiple choice
- Random
- Population of 3 roles: Main (general strong) / Main Exploiters (target Mains) / League Exploiters (cover blind spots); mixture of past + current opponents prevents cycling; Grandmaster StarCraft II
- Same as self-play
- Not real
AlphaStar league: Mains + Main Exploiters + League Exploiters; StarCraft II Grandmaster.
#multi-agent
Practise Reinforcement Learning
214 interview questions in this topic.