Blizzar and DeepMind share StarCraft II machine learning tools

Posted on Thursday, August 10 2017 @ 13:51 CEST by Thomas De Maesschalck
Last year Google's DeepMind unit announced a collaboration with game developer Blizzard to train artificial intelligence (AI) by testing agents in StarCraft II. Now the companies share the results of their work by making the StarCraft II machine learning tools available to the general public:
Testing our agents in games that are not specifically designed for AI research, and where humans play well, is crucial to benchmark agent performance. That is why we, along with our partner Blizzard Entertainment, are excited to announce the release of SC2LE, a set of tools that we hope will accelerate AI research in the real-time strategy game StarCraft II. The SC2LE release includes:

  • A Machine Learning API developed by Blizzard that gives researchers and developers hooks into the game. This includes the release of tools for Linux for the first time.
  • A dataset of anonymised game replays, which will increase from 65k to more than half a million in the coming weeks.
  • An open source version of DeepMind’s toolset, PySC2, to allow researchers to easily use Blizzard’s feature-layer API with their agents.
  • A series of simple RL mini-games to allow researchers to test the performance of agents on specific tasks.
  • A joint paper that outlines the environment, and reports initial baseline results on the mini-games, supervised learning from replays, and the full 1v1 ladder game against the built-in AI.
  • Below is a video clip that illustrates how an untrained and a trained agent handle a StarCraft II mini game:



    You can check it out at DeepMind. The research unit says they've managed to train AI agents that excel at performing very specific tasks like moving the camera or collecting minerals. But the creation of an AI that can play StarCraft II is still in its infancy. At the moment, even the strongest AI agents fail to beat the game's easiest, scripted AI.

    DeepMind hopes to improve the StarCraft II machine learning agents by using imitation learning. Blizzard will provide the researchers with hundreds of thousands of anonymized replays from the StarCraft II ladder to train the supervised agents to get better.


    About the Author

    Thomas De Maesschalck

    Thomas has been messing with computer since early childhood and firmly believes the Internet is the best thing since sliced bread. Enjoys playing with new tech, is fascinated by science, and passionate about financial markets. When not behind a computer, he can be found with running shoes on or lifting heavy weights in the weight room.



    Loading Comments