Google's AI wins final Go challenge

Google's DeepMind artificial intelligence has secured its fourth win over a master player, in the final of a five match challenge.

Lee Se-dol, one of the world's top Go players, won just one of the matches against the AlphaGo program, missing out on the $1m prize up for grabs.

Demis Hassabis, founder of DeepMind, said the match had been the "most exciting and stressful" for his team.

Lee Se-dol said he felt "regrettable" about the result of the contest.

Mr Lee won one of the five matches

In Go, players take turn placing stones on a 19-by-19 grid, competing to take control of the most territory.

The game is considered to be much more challenging for computers than chess.

At a press conference held after the final match, Mr Lee said he did not necessarily think AlphaGo was superior to humans.

But he said he had more studying to do, and admitted the matches had challenged some of his ideas about the game Go.

Analysis: Stephen Evans, Seoul correspondent

The competition has been seen as a landmark moment for artificial intelligence

In some countries the people watch football on big screens in public squares, but in South Korea it's been the mighty challenge of machine against humanity.

And the victory of the computer has led to some introspection.

One South Korean newspaper complained that the contest was "lopsided" with the single Korean pitted against the corporate might of Google and its "army of super-smart people armed with unfathomable computing power".

In a spirit of magnanimity, however, the Korea Baduk Association - which governs the game of Go - has decided to give an honorary ninth-dan ranking to AlphaGo.

The 4-1 mechanical victory has also made some Go players doubt themselves.

The European champion who lost last year to AlphaGo said it had really knocked his self-confidence, even as it enabled him to climb up the world rankings.

Go was invented about 2,500 years or so ago in China. Until now, it has always had a human best player. Not any more.

The five match challenge began in Seoul on 9 March, where AlphaGo scored its first victory.

After losing the second match, Lee Se-dol said he was "speechless" adding that the AlphaGo machine played a "nearly perfect game".

In the third game commentators said that Lee Se-dol had brought his "top game" but that AlphaGo had won "in great style".

DeepMind's winning streak meant it won the $1m (£702,000) prize on offer. Google said the money would be donated to Unicef, Stem (science, technology, engineering, and maths) charities and Go organisations.

Mr Hassabis said: "We have been lucky to witness the incredible culture and excitement surrounding Go.

"Despite being one of the oldest games in existence, Go this week captured the public's attention across Asia and the world."

Expertise

The AlphaGo system was developed by British computer company DeepMind which was bought by Google in 2014.

It has built up its expertise by studying older games and teasing out patterns of play.

Lee Se-dol did win the fourth match against AlphaGo, after which he said: "I've never been congratulated so much because I've won one game."

Despite Mr Lee's overall defeat, rival players have still expressed confidence that they could beat the AI.

Ahead of the final, China's top ranked Go master Ke Jie said he believed he could beat AlphaGo.

He told China Central Television: "In terms of probability, I have a chance to win, but the probability is not as high as I thought before. I think it is 60 per cent in favour of me."

Analysis by Dr Noel Sharkey, AI expert

To beat one of the world's top players, Deep Mind used a mixture of clever strategies to make the search much smaller.

Does this mean AI is now smarter than us and will kill us mere humans? Certainly not.

AlphaGo doesn't care if it wins or loses. It doesn't even care if it plays and it certainly couldn't make you a cup of tea after the game.

Does it mean that AI will soon take your job? Possibly you should be more worried about that.

What is Go?

Go is thought to date back to several thousand years ago in China.

Using black-and-white stones on a grid, players gain the upper hand by surrounding their opponents pieces with their own.

The rules are simpler than those of chess, but a player typically has a choice of 200 moves, compared with about 20 in chess - there are more possible positions in Go than atoms in the universe, according to DeepMind's team.

It can be very difficult to determine who is winning, and many of the top human players rely on instinct.