Google’s DeepMind AlphaGo artificial intelligence (AI) managed to beat the world’s top Go player Ke Jei.
The AI agent defeated Jei in a three-part match after securing a win in the second round.
In a tweet posted while the match was occuring DeepMind founder and CEO, Demis Hassabis, said: “Incredible. According to #AlphaGo evaluations Ke Jie is playing perfectly at the moment.”
Incredible. According to #AlphaGo evaluations Ke Jie is playing perfectly at the moment.
— Demis Hassabis (@demishassabis) May 25, 2017
After the match was over Hassabis said Jie managed to push the AI “right to the limit”.
#AlphaGo wins game 2. What an amazing and complex game! Ke Jie pushed AlphaGo right to the limit.
— Demis Hassabis (@demishassabis) May 25, 2017
In the post-match press conference, Ke Jie said: “If you know Go, you know that AlphaGo in today’s game, played some moves which were opposite from his vision of maximising the probability of winning. From the perspectives of human beings, he stretched a little bit and I was surprised. I also thought I was very close to winning the match in the middle of the game, but that might not have been what AlphaGo was thinking. I was very excited, I could feel my heart thumping!”
Demis Hassabis, CEO and co-founder of DeepMind, commented: “The first 100 moves were the best anyone’s ever played against the Master version. Our hearts were fluttering as well! I’d like to reiterate what an honor it is to play with a genius like Ke Jie. This is called the Future of Go Summit, and today I think we saw a game from the future.”
Kim Sung Yong, 9 Dan Professional and commentator from Japan, said: “AlphaGo showed the definition of ‘mind-blowing’ play today. When human artists start drawing landscapes, they keep drawing landscape no matter what happens during the process. However, AlphaGo can quickly switch from landscape to portrait. Similarly, human players have their own style of Go plays. However, AlphaGo has no fixed style and is so flexible.”
Last year AlphaGo bested another top Go player, Lee Sedol, winning four out of five matches. In October 2015, the AI agent defeated European champion Fan Hui 5-0.
“One of the great promises of AI is its potential to help us unearth new knowledge in complex domains. We’ve already seen exciting glimpses of this, when our algorithms found ways to dramatically improve energy use in data centres – as well as of course with our program AlphaGo,” DeepMind says on its website.
The AI agent is able to make excellent use of “influence,” or the effect of existing stones on surrounding areas. While influence cannot be measured exactly, DeepMind says that its “value network enables it to consider all stones on the board at once.”
