Science

Blizzard and DeepMind turn StarCraft II into an AI research lab

Blizzard and DeepMind turn StarCraft II into an AI research lab

To help test the AI, there are a few simple mini-games included that work within the bounds of vanilla StarCraft II, though researchers that want to use mods or create their own mini-games can easily do so using PySC2 and Blizzard's machine learning API. Starcraft is considered an important target for machine learning researchers because, unlike Go, in which both players can see the entire board and take turns moving pieces, players in Starcraft cannot see what is happening in the entire game environment at one time and both players move their units simultaneously.

Google's artificial intelligence (AI) lab DeepMind and game development studio Blizzard have announced the release of a set of tools aimed at accelerating AI research through real-time strategy game StarCraft II. In partnership with Blizzard, programmable A.I. agents are being implemented into the Starcraft II API.

For example, while the objective of the game is to beat the opponent, the player must also carry out and balance a number of sub-goals, such as gathering resources or building structures. The built-in agents, which are created by Starcraftpublisher Activision Blizzard, use hard-coded rules rather than the advanced machine learning techniques DeepMind specialises in.

"In addition, a game can take from a few minutes to one hour to complete, meaning actions taken early in the game may not pay-off for a long time".

More news: Modi ruined Gujarat, now he's ruining country: Ahmed Patel
More news: 2016 was Earth's warmest year on record, worldwide report shows
More news: Sanchez to miss start of season for Arsenal

StarCraft II is such a useful environment for AI research basically because of how complex and varied the games can be, with multiple open routes to victory for each individual match. This means that not only does the A.I. need to learn how to play the game itself, it needs to recognize general strategy, map variation, and predictive tactical reasoning.

DeepMind announced today that it is releasing the Starcraft II Learning Environment (SC2LE), a toolset that makes it easy for AI developers to use Starcraft II to test their AI.

Although the AI "agents" are able to perform tasks well in isolated mini-games, when exposed to the full game they struggle with tasks that humans find fairly trivial - it's this that Deepmind hopes to combat with Blizzard's help. DeepMind has published a number of research papers that hint it may be closing in on creating software capable of numerous tasks, such as prioritising goals, long-term planning, and memory that any system will need to play Starcraft IIsuccessfully.

According to a blog post by DeepMind research scientist Oriol Vinyals, program manager Stephen Gaffney, and software engineer Timo Ewalds, testing AI in games that were not designed for such research and in which human players excel "is crucial to benchmark agent performance".