Science | Artificial Intelligence
Meet Leela zero a chess artificial intelligence that works on a self-taught neural network and general reinforcement learning algorithm.
It can make smart and lethal moves on its own.
The network learns via deep learning techniques by playing against itself a million times.
No one can defeat it, not even the world champion
DeepMind published a paper that narrates an Artificial Intelligence for playing games.
The researchers had advanced it for two-person games, peculiarly for Go, chess, and shog (“which is Japanese chess.”)
It was given no particular information beyond the rules of each game. First, it made random moves, then it started learning by itself.
Surprisingly, things got excited when it started performing marvelous than normal human players; after 4 hours, it was crushing the best chess engine in the world.
Over the span of 9 hours, the chess system played 44 million games against itself on specialized Google hardware.
Isn’t it mind-boggling?
The program called Alpha zero, that became famous for defeating the world’s best Go players.
The neural network of Alpha zero works in systematic ways, the network receives input on the layout, several recent moves of the game.
As output, it evaluates the current player’s possible ways to winning and predicts which of the move are likely to work best.
Chess has been used as a Rosetta Stone of both human and machine cognition for over a century. AlphaZero renews the remarkable connection between an ancient board game and cutting-edge science by doing something extraordinary.”(Garry Kasparov, the former chess champion)
A journey toward intelligent and self-aware Artificial Intelligence. It has extraordinary potential to benefit Human beings. We just need to be responsible for the collective good.