

9 years ago
2
Gamasutra: Ben Weber's Blog - DeepMind Challenges for StarCraft
AlphaGo recently played against 9-dan professional Go player Lee Sedol. The AI won the first three games against the human opponent, achieving victory in the best-of-five tournament. With this challenge accomplished, the DeepMind team is looking for new problems to use as a testbed for the system. Demis Hassabis, co-founder of DeepMind, expressed interest in StarCraft as a challenge.
Continue Reading
Join the Discussion
Would it really be a challenge? APM, multi-tasking, and micro-managing would mean nothing to a machine. A machine could probably do scouting like no human has ever done before. Maybe a more interesting challenge would be a group of different and independent (unconnected) AIs trying to play as a team on a game like League of Legends or Dota2.
Chess and go were fascinating challenges because they give an opportunity to humans think on these classic over-a-thousand-years-old turn-based games.
Interesting how these games are probably landmarks in the history of the evolution of AI.
I totally agree that Starcraft II might not be a great benchmark. Between perfect multitasking and a huge micro advantage, an AI with imperfect macro strategies could still be made very, very hard to beat.