Hi Charlie !

Today I am going to write an article about the Google creation of an artifical intelligence, clever enough to win four times on five against the better player of "go" of the world .
What is the play of go ?
The play of go is a game from China in a chaquered tray . Two players are opposed, and have to put little stones (respectively black or white) on the tray. The goal is to control the tray in building "territories", circled the stone of the oppenent and make them"prisonner". The winner is the one who made the most of territories and prisonners.
We need to play this game a huge sens of logic and strategy.

On Tuesday, 15th of March, the player created by Google " DeepMind AlphaGo " won in five rounds the better go player in the world, the Korean, Lee Sedol, with the score of 4-1 for AlphaGo.

This the first round.
The video is really long and it really begins after 30 minutes.


After the last round, Lee Sedol was interrogated. He said "Actually , AlphaGo is different, first of all because he is not human. Evertything was completly different from what I used to know. The adaptation was complicated to me."

Lee Sedol also won a round, prooving that AlphaGo is not infallible. When the korean executed his 78th action (the decisive one), Demis Hassad, the creator of AlphaGo told us how the artificial intelligence "lived" it.
He said " I quicly checked his internet registers : AlphaGo estimated less than 1 on 10 000 the probability of the brilliant Lee's 78th action. AlphaGo was really surprised."

AlphaGo made also big mistakes. Hassad was really proud of his DeepMind AlphaGo, but know that he has to ameliorate the algoritmes to improve the artificial intelligence of his machine.
Lee Sedol has also to work on new strategies. He said " The old beliefs of the human creativity and our traditional knowledge had been challenged by my experience. I realize that I have work to do in this way".


Be yourself,
Amma