Go champion finally wins against Google's AlphaGo program

14 Mar 2016 | By Shiladitya

Top echelon South Korean Go player Lee Sedol won his first match against an artificial intelligence program called AlphaGo deveploped by DeepMind, one of Google's subsidiaries.

AlphaGo made a mistake on move 79 which cost it the match.

This was the 4th match in a 5 match series between AlphaGo and Lee Sedol; Sedol had lost the last 3 games.

In context: Man vs Machine: Human players humbled by AI

Go The game of Go

The game of Go originated in China more than 2,500 years back.

Two players alternately place black and white stones with an aim of surrounding more territory than his opponent in a board of 19x19 grid lines.

The game is exponentially more complex than chess - the number of possible games is 10^761 in Go as opposed to Chess' estimated 10^120 possible games.

AI programs fail to master Go

Computer programs mastered several games over the years beating their human competition - Tic-tac-toe in 1952, Checkers in 1994, Chess in 1997, and Jeopardy! in 2011. However, they could only play Go at an amateur level till AlphaGo's advent.
Love Tech news?
Stay updated with the latest happenings.
Google subsidiary DeepMind's AlphaGo program

AlphaGo Google subsidiary DeepMind's AlphaGo program

Google owned DeepMind's AlphaGo was an artificial intelligence program developed in London to play the ancient Chinese board game Go.

AlphaGo displayed a marked improvement from all other existing Go-playing programs.

In 500 games against all other available Go programs, AlphaGo running on a single computer won all but one.

Running on multiple computers, AlphaGo won all 500 matches against other Go programs.

October 2015DeepMind's AlphaGo makes history

AlphaGo became the first computer program in history to beat a professional level human Go player in a full sized board without any handicaps.

AlphaGo dominated the reigning three-time European Go champion, Fan Hui, a 2 dan (out of a possible 9) professional, beating the human 5-0 in a 5 match series.

AlphaGo's next opponent was to be the legendary Go champion, Lee Sedol.

Lee Sedol's legendary status

Lee Sedol, born in 1983, is a professional Go player with a 9-dan rank - the highest possible. As of February 2016, he ranked second in the world in international titles (18), behind only Lee Chang-ho (21).

12 Mar 2016AlphaGo clinches series against Lee Sedol

AlphaGo won its third consecutive match against legendary 9 dan Go professional, Lee Sedol, clinching the 5 match series 3-0.

Sedol apologized to his followers for his inability to beat the machine.

He said, "I do apologize for not being able to satisfy a lot of people's expectations. I kind of felt powerless", adding that he underestimated the machine's capabilities.

Love Tech news?
Stay updated with the latest happenings.

14 Mar 2016Go champion finally wins against Google's AlphaGo program