Go-playing AI or better known as AlphaGo represents a brand new concept of what we thought it was artificial intelligence. The AI is based on a learning machine but this one is specialized to play the ancient game of ‘Go’ a task a bit difficult for a machine to learn due to the limited amount of branching paths that can be done in the game.

After a win streak against other humans and even national champions and other AI’s, it was announced that the machine needed a new challenge but this time, they would not defy another regular human or just another AI but the current Go World champion, Lee Sedol from Korea, they would play in a series of planned matches. Earlier today, the AI defeated Sedol on the first of five matches.

Traditional AI’s tactics such as brute-forcing resulted in being somehow ineffective but considering that AlphaGo possesses the newer technology, it does not present those ineffective tactics anymore. Credit: laughing squid

Human vs Machine

It all started with a polite reverence before the match, such gesture may not have meant too much to AlphaGo but on the other hand, to Sedol, it meant he was standing before a worthy opponent. After three hours and a half later, the epic game between human and machine concluded and as nobody expected it, AlphaGo ended up victorious.

Indeed, it was a rough match, both players made some mistakes during the game but as any other human being with feelings Seol got nervous, forgot a couple of key moves while taking quick sips of water, something that AlphaGo did not suffer from, it never hesitated because, after all, it is a machine, most of the mistakes the AI made were product of its nature as a machine that is for certain less capable of rational thought which is a fundamental perk in the complex game of Go.

In a press conference after the match Sedol seemed defeated and utterly destroyed as if his whole world crumbled, the reporters described him as being “on the verge of tears” as he talked about the match, Seol said, “I didn’t know the AlphaGo would play such a perfect game.”

This event was stunning for the whole world of Go and its community, many followers, players and enthusiasts thought that the machine was going to be defeated by the current world champion, the best human player that the game had to offer but as nobody thought the results were the opposite.

The AlphaGo A.I used its “Reinforcement learning” programming to learn from its mistakes in matches against other A.I’s and other humans as well as against itself.

At this point, Sedol probably does not stand a chance to win any of the four remaining games.

Source: Android Headlines