Wednesday, March 9, 2016

Google Software defeat human champion Go – Azteca News

Seoul, Korea Sur.- The artificial intelligence Google AlphaGo Wednesday defeated his rival human , South Korean champion Lee Sedol in the first game of Go of its historical confrontation to five games.

the victory of AlphaGo in this ancient Chinese board game is an improvement for the artificial intelligence , since it demonstrates that the program developed by Google DeepMind has mastered one of the most creative complex games ever created.

the experts they noted that the game was tight and both AlphaGo as Lee made some mistakes, with an unpredictable outcome until he was near the end.

the defeat of Lee was a blow to the South Koreans and fans go. The 33-year-old was initially relied on a strong victory two weeks ago, but the day before the game was less optimistic.

“I was very surprised because I did not think it was to lose game. a mistake I made at the beginning lasted until the end, “Lee, who has won 18 world championships since he turned pro at 12, said.

Lee said the strategy of AlphaGo had been “excellent” and that had caused a great impact unconventional machine movement, a human never would have used.

Despite his initial defeat, Lee said not regret having accepted the challenge.

“I had fun playing go and I’m looking forward to play new games,” he said with a quiet smile.

the result shocked the South Korean community amateur Go .

“He did not play like a human at all,” said Yoo Chang-hyuk, also master Go South Korea on the lack of emotion of the machine, which however made some mistakes that could have cost him the game.

Hundreds of thousands of people watched the game live on television. The remaining four games will end on Tuesday.

Background of artificial intelligence

Computers dominated chess in 1997 in a game between Deep Blue, IBM, and chess champion Garry Kasparov, leaving the Go as “the only game that is above chess,” he said before the game Wednesday Demis Hassabis, CEO of Google DeepMind.

the big players Go depend on your intuition and emotions to make their moves to choose from an almost infinite possibilities on the board, making the game a challenge for developer community artificial intelligence.

experts in artificial intelligence had predicted would happen another decade before computers could beat professional players Go until AlphaGo defeated European champion game last year in a game behind closed doors which are then reported in the journal Nature. Since then, the ability to AlphaGo has improved steadily.

His victory over a human indicates that computers can mimic intuition and tackle more complex tasks, said its creators. That capacity, they said, could be used to help scientists solve difficult problems in the real world, in areas such as health.

LikeTweet

No comments:

Post a Comment