Computers have been able to beat humans at Chess for a while, but until now, they haven’t come close to beating professional Go players because of all of the possible combinations in the game. But now, in a major breakthrough, Google has developed AlphaGo, an AI that beat the European champion, Fan Hui, five games to none.
Since Go is a game that has a lot of intuition and feel, computers are generally worse than the professional human players. In order to make AlphaGo competitive against professionals, the software analyzed millions of moves from games played by humans. It learns patterns, and through analyzing and playing games it gets better and can have more intuitive knowledge about what positions are good. This is a great breakthrough for AI and machine learning.
In March, Google will test AlphaGo against Lee Sedol, the world’s top Go player.
For more information, read the full article here.