Google’s AlphaGo defeats human in first match
South Korean Go player, Lee Se-dol is competing against Google’s artificial intelligence software, AlphaGo
Google has developed the first artificial intelligence software, AlphaGo, to have mastered the art of playing Chinese board game, Go, and is now competing in a five-game match against a human in Seoul, South Korea.
The AI, developed by Google’s DeepMind unit, today defeated legendary Go player Lee Se-dol, in the first of five matches to be played. Lee resigned after three and a half hours, with 28 minutes and 28 seconds remaining on his clock.
"I was very surprised, I didn't expect to lose. [But] I didn't think AlphaGo would play the game in such a perfect manner,” said Lee after the match. "I don’t regret accepting this challenge. I am in shock, I admit that, but what's done is done. I enjoyed this game and look forward to the next. I think I failed on the opening layout so if I do a better job on the opening aspect I think I will be able to increase my probability of winning."
DeepMind claims its AlphaGo search algorithm can predict which moves it thought will most likely win. Already, AlphaGo won 5-0 in 2015 against three-time European Go champion Fan Hui, this win encouraged DeepMind researched to play against Lee. The South Korean player has a history of winning Go tournaments in South Korea, Japan, China and Taiwan.
The series is the first time a professional 9-dan Go player has taken on a computer. Competing for a $1m prize, Lee will face AlphaGo again tomorrow, and on Saturday, Sunday and Tuesday.
Go was invented over 2,500 years ago. It typically consists of a 19-by-19 square board, where players use black and white stones to attempt to capture empty areas and surround their opponent's stones. The complex game can be likened to chess, but where chess offers 20 possible choices per moves, Go provides the player with 200.