Google DeepMind AI program defeats S. Korean Go master

Google DeepMind AI program AlphaGo defeated human Go world champion, Lee Sedol, in a five game series. The victory draws the week-long Challenge Series tournament in South Korea to a conclusion.

This win marks a major milestone for artificial intelligence research, as Go is a fairly simple game, but has proven to be infamously difficult for computers to master due to the vast number of potential moves. Many Go players say the game primarily relies on intuition as a strategy.

The victory is also significant, as it marks the first time in history that an AI program beat one of the best Go players in the world. The game itself is a two-player, turn-based strategic competition. Each player either puts down black or white stones in an attempt to outsmart and surround the other player. Go is rather simple to pick up on, but it takes years to master.

AlphaGo established itself early on, winning three straight matches in the five-game series. Sedol rebounded by winning the fourth match after pushing the AI to make a mistake on move number 79. He resigned for the fifth match, giving AlphaGo the final game and a 4-1 overall victory.

For the first time in the tournament, Sedol played as black to try to confuse the AlphaGo AI, but ultimately lost. He noted he was “regretful” of the result and added that he has “more studying to do.”

“AlphaGo made what looked like a mistake with move 48, similar to what happened in game four in the middle of the board. After that AlphaGo played very well in the middle of the board, and the game developed into a long, very difficult end game,” Michael Redmond, English commentator and Go professional, said.

The tournament was closely examined by many senior people at Alphabet and Google. Alphabet president Sergey Brin attended the third game and Alphabet chairman Eric Schmidt was there for the first.

After the third game, Sedol apologized to fans for not being able to satisfy their expectations. “I kind of felt powerless,” he said. “When I look back on the three matches, even if I were to go back and redo the first match, I think I would not be able to win because I misjudged AlphaGo.”

The victory poses a question we’ve been wondering for quite some time: If artificial intelligence can outsmart humans, what does the future hold?

Source: Business Insider