Advertise on ITWeb         Mon, 29 Aug, 10:50:25 AM

Google's AlphaGo wins four out of five

Go world champion Lee Sedol places the first piece against Google's artificial intelligence program AphaGo.

Go world champion Lee Sedol places the first piece against Google's artificial intelligence program AphaGo.

Over 70 000 people tuned in to YouTube this morning to watch Google's artificial intelligence program beat world Go champion Lee Sedol, again.

This was the final match of five in the Google DeepMind Challenge taking place in Seoul, South Korea.

The victory for the AlphaGo program, designed by Google subsidiary DeepMind, over Lee, the holder of 18 international titles, surprised many, including its designers, and underscored advances in artificial intelligence.

Lee won one match out of five: the fourth match that took place on Sunday.

Experts did not expect an artificial intelligence program to beat a human professional for at least a decade, until AlphaGo beat a player last year. But Lee, 33, had been expected to pose stiffer competition than the player defeated then.

"To be honest, we are a bit stunned and speechless," said DeepMind founder Demis Hassabis after the third match, earlier tweeting that AlphaGo's victory was a "historic moment".

Go is an ancient Chinese board game, said to be much harder than Chess. It involves two contestants moving black and white stones on a square grid.

The British Go Association says Chess is a hierarchical game and Go an imperial game. The object of Chess is to catch the king and the object of Go is to enclose more territory on the board than the opponent does.

"At the opening move in Chess, there are 20 possible moves," states the association. "In Go, the first player has 361 possible moves. This wide latitude of choice continues throughout the game. At each move the opposing player is more likely than not to be surprised at his opponent's move, and hence he must rethink his own plan of attack."

Google executives say Go offers too many possible moves for a machine to win simply through brute-force calculations, unlike Chess, in which IBM's Deep Blue famously beat former world champion Garry Kasparov in 1997.

Instead, they said, AlphaGo sought to approximate human intuition, by studying old matches and using simulated games to hone itself independently.


Our comments policy does not allow anonymous postings. Read the policy here




 

 

 

Sponsors Message