AlphaGo, Google AI Program Defeats Go Legend Lee Se-dol For The Second Time
Google is once again making the news, this time because of a great feat of its AlphaGo beating legendary Go player Lee Se-dol not once, but twice in a series of five historic matches that is currently being held in Seoul, South Korea. It is one the tech giant’s program developed under its DeepMind unit for its artificial intelligence business side.Advertisement
Just this Wednesday, The Verge reported that the first of the five series games was done and ended up with the professional 9-dan Go player resigning after about three and a half hours into the game with a remaining 28 minutes and 28 seconds in the clock. The match left a lot surprised, stunned, victorious and doubtful. Surprised and stunned that an AI has defeated a person for the first time it seems.
The DeepMind team was victorious, as their “creation” has managed to achieve such an astounding feat. And like always, doubtful for those who think the achievement was a fluke, after all the game was still the first of the five that is happening between the two competitors.
After the match Lee said that he was very surprised. He did not expect to lose in the game and he definitely didn’t think that AlphaGo will play the game in such a perfect manner.
The following day, Google’s creation has once again stunned the world and proved that its victory over Lee wasn’t a fluke – it won again! The intense battle this time saw Lee resigning after the final period of “byo-yomi” or the second reading overtime which gave him about 60 seconds to carry out each move.
The Verge says Lee was left speechless. He was heard saying on the post game conference that yesterday he was surprised but “today I am speechless. I admit that it was a very clear loss on my part. From the very beginning of the game I did not feel like there was a point that I was leading.”
Meanwhile DeepMind founder, Demis Hassabis, who said that he has a huge respect for Lee and his amazing skills was ecstatic and speechless too. He said that he think it’s a “testament to Lee Se-dol’s incredible skills. We’re very pleased that AlphaGo played some quite surprising and beautiful moves, according to the commentators, which was amazing to see.”
The turn of events seems to offer validation of the AlphaGo’s evaluative ability, which in the previous Go programs has been the main problem in its proficiency. And until Wednesday, the game has never been played on a world class level by an AI. It was also the first time that a professional 9-dan Go player has taken on a computer with a $1 million USD prize at stake.
However, it is not the first time that the AI has played against a person on a Go match. Last November it could be recalled that the program has played against European Go champion Fan Hui. Hui has been an elite professional Go player who devoted his life to Go since he was 12 years old. In the closed doors match, DeepMind’s AI program has won all five games reports Slate.
The third match according to The Verge reports will be this Saturday, March 12 followed by the fourth game the next day, March 13. The final game is set for Tuesday, March 15. Each game which is expected to last from four to five hours will be live streamed on DeepMind’s YouTube so if you’re interested you can watch it there. After each game, updates by the tech giant will also be posted on its Asia Pacific blog.
Also Read: Facebook OCP’s Newest Contributor Is Google
For more technology news, you can subscribe to our newsletter. If you are a social media junkie and would like to receive updates on your timeline, then you can also like us on Facebook or follow us on Twitter.