Gaming Google DeepMind could invent the next generation of AI by playing Starcraft 2

tom_mai78101

The Helper Connoisseur / Ex-MineCraft Host
Staff member
Reaction score
1,678
The announcement at BlizzCon 2016 that met with the most muted response was arguably the most revolutionary.

While new content for the likes of Hearthstone, Heroes of the Storm, Overwatch, and Diablo III drew appreciative roars from the Blizzard faithful, the news that Google’s DeepMind branch—which is dedicated to developing sophisticated Intelligence—would be teaming up with the makers of Starcraft 2 to further its research on AI elicited more of a murmur.

Perhaps the lack of enthusiasm was down to taste. After all, why would the plans of AI scientists be of interest to Starcraft 2 players? As it turns out, if the collaboration between DeepMind and Blizzard is what its developers hope it could be, players will see very tangible benefits—and so will many others outside the video game space.


AI has been an integral part of video games for quite some time now, and as Starcraft 2’s executive producer Chris Sigaty points out, the AI that DeepMind hopes to develop won’t be the first to play the Blizzard RTS. Currently the games uses a scripted AI. It's created by a team of designers and engineers that have an understanding of the way the game plays, with their expertise placed into a script and fed a list of conditions. For example, if X amount of resources is collected, the AI should spend the value of Y on Z units.

"They’re elaborate and interesting to play against; there’s some random choice and specific choices that happen," says Sigaty. "They can expand on their terrain, gather resources—basically do the things you need to do in Starcraft 2. But these [AI] are scripted. They’re predictable and ultimately, unless it cheats, there’s not an AI out there who can beat a human."

This holds true for Starcraft 2—the best e-sports players routinely beat AI opponents. The aim DeepMind is to develop an AI that plays the game the same way as a human, potentially one that can beat the best players, which widens the potential for deep learning. And the DeepMind team has already had some success in the gaming field: in March this year, the team’s AI AlphaGo saw off the world-class Go player Lee Sedol in a best-of-five series that it won 4-1—leaping what was a major hurdle for AIs in the past.


This is a followup to this news article.
 
Last edited by a moderator:

Ghan

Administrator - Servers are fun
Staff member
Reaction score
888
This will be interesting to see. The leap from perfect information games to partial information games is not trivial. I'll be interested to see what they come up with.
 

Accname

2D-Graphics enthusiast
Reaction score
1,462
This is going to fail. Not in a big way though. I bet they will create an AI which is okay, but I highly doubt it will be great. It will simply be okay and that can be considered a fail for a project like this.
 

seph ir oth

Mod'n Dat News Jon
Reaction score
262
Given how well neural networking can teach an AI how to play, I think it's quite possible this will succeed.

 

Accname

2D-Graphics enthusiast
Reaction score
1,462
Nah, its all garbage.
Maybe in 30 - 50 years. But right now this will not become anything special.
 
General chit-chat
Help Users
  • No one is chatting at the moment.

      The Helper Discord

      Members online

      Affiliates

      Hive Workshop NUON Dome World Editor Tutorials

      Network Sponsors

      Apex Steel Pipe - Buys and sells Steel Pipe.
      Top