13 votes

DeepMind's StarCraft II AI "AlphaStar" has reached the top-level Grandmaster league using multi-agent reinforcement learning

2 comments

  1. skybrian
    Link
    From the article:

    From the article:

    The key insight of the League is that playing to win is insufficient: instead, we need both main agents whose goal is to win versus everyone, and also exploiter agents that focus on helping the main agent grow stronger by exposing its flaws, rather than maximising their own win rate against all players. Using this training method, the League learns all its complex StarCraft II strategy in an end-to-end, fully automated fashion.

    2 votes
  2. joelthelion
    Link
    I wonder if they will continue working on it until they crush all humans, or if this is it? Did they say what their next goals might be?

    I wonder if they will continue working on it until they crush all humans, or if this is it? Did they say what their next goals might be?