DeepMind Research on Ladder

Experimental versions of DeepMind’s StarCraft II agent, AlphaStar, will soon play a small number of games on the competitive ladder in Europe as part of ongoing research into AI.

If you would like the chance to help DeepMind with its research by matching against AlphaStar, you can opt in by clicking the “opt-in” button on the in-game popup window. You can alter your opt-in selection at any time by using the “DeepMind opt-in” button on the 1v1 Versus menu.

For scientific test purposes, DeepMind will be benchmarking AlphaStar’s performance by playing anonymously during a series of blind trial matches. This means the StarCraft community will not know which matches AlphaStar is playing, to help ensure that all games are played under the same conditions. AlphaStar plays with built-in restrictions that the DeepMind team has defined in consultation with pro players. A win or a loss against AlphaStar will affect your MMR as normal.

The FAQ below reflects the team’s plans for AlphaStar as of the time of posting, but may be subject to change as their research goes on.

Q. Why is AlphaStar available to play on Battle.net?

A. Experimental versions of DeepMind's StarCraft II system, AlphaStar, will soon play a small number of games on Battle.net for a limited time as part of ongoing scientific research into artificial intelligence. We’re excited to offer our community a chance to contribute to this process by playing against AlphaStar in a small number of blind trial games that will allow DeepMind to benchmark its agents.

Q. What version of StarCraft II is AlphaStar playing?

A. AlphaStar will play on the latest version of StarCraft II.

Q. What races can AlphaStar play as?

A. AlphaStar can play as and versus Terran, Zerg or Protoss.

Q. Why is AlphaStar playing anonymously?

A. DeepMind is currently interested in assessing AlphaStar’s performance in matches where players use their usual mix of strategies. Having AlphaStar play anonymously helps ensure that it is a controlled test, so that the experimental versions of the agent experience gameplay as close to a normal 1v1 ladder match as possible. It also helps ensure all games are played under the same conditions from match to match. DeepMind will release the research results in a peer-reviewed scientific paper along with replays of AlphaStar's matches.

Q. How will I be able to play against AlphaStar?

A. AlphaStar will play anonymously during a series of blind trial matches against players on the competitive ladder. Players will be paired against AlphaStar according to the normal matchmaking rules. You can change your preference to opt in or opt out at any time via the Versus screen.

Q. How many variants of AlphaStar will play?

A. DeepMind will be benchmarking the performance of a number of experimental versions of AlphaStar to enable DeepMind to gather a broad set of results during the testing period.

Q. How likely am I to be matched against AlphaStar?

A. Pairings on the ladder will be decided according to normal matchmaking rules, which depend on how many players are online while AlphaStar is playing. AlphaStar will play a small number of games for scientific test purposes, though we will not be revealing exactly when or how often this will happen to ensure that games remain anonymous.

Q. How does AlphaStar perceive the game?

A. Like human players, AlphaStar perceives the game using a camera-like view. This means that AlphaStar doesn’t receive information about its opponent unless it is within the camera’s field of view, and it can only move units to locations within its view. All limits on AlphaStar’s performance were designed in consultation with pro players.

Q. How does AlphaStar interact with the game?

A. AlphaStar has built-in restrictions, which cap its effective actions per minute and per second. These caps, including the agents’ peak APM, are more restrictive than DeepMind’s demonstration matches back in January, and have been applied in consultation with pro players.

Q. Will AlphaStar play team matches or only 1v1?

A. AlphaStar will only be playing 1v1 during these test matches.

Q. What regions will AlphaStar play in?

A. Currently, AlphaStar agents will play in Europe.

Q. Will playing AlphaStar affect my MMR?

A. A win or a loss against AlphaStar will affect your MMR like any other game played on the ladder.

Q. Will AlphaStar improve as it plays on the ladder? Will my games be used to help improve its strategy?

A. AlphaStar will not be learning from the games it plays on the ladder, as DeepMind is not using these matches as part of AlphaStar’s training. To date, AlphaStar has been trained from human replays and self-play, not from matches against human players.

Q. What’s happening next with AlphaStar?

A. Once AlphaStar has played enough test matches, the team at DeepMind will use the results to inform their ongoing research into artificial intelligence. They will release the research results in a peer-reviewed scientific paper along with replays of AlphaStar’s matches, and are working with us to explore what comes next for AlphaStar. If DeepMind has any other updates on AlphaStar, the community will be the first to know!

Q. I pressed the wrong button and opted in/opted out by accident. Can I change my preference?

A. If you change your mind about playing against AlphaStar, you can change your preference to opt in or opt out at any time via the Versus screen.

Q. How will my replays against AlphaStar be used?

A. DeepMind will use your replays and the game data they include as part of their scientific research assessing and describing the performance of the AlphaStar system. In particular, replays will help illustrate specific points about AlphaStar’s gameplay. For more information, please see the privacy policy.

Q. I have a question about how DeepMind is processing my personal data. What should I do?

A. You can contact DeepMind by emailing replaydataqueries@deepmind.com. Please also see the privacy policy for more information.