Researchers presented AlphaZero, a computer program capable not only of achieving super-human mastery in some of the most sophisticated games on board, such as Chess, Shogi and Go but also learning to play without prior knowledge. , except for the rules for each game. "
The system was created by experts from DeepMind, a British company owned by Alphabet, the parent company of Google.
The results, the researchers said, represent an important step towards the development of an AI player who can quickly learn and master each game.
Chess, Shogi and Go
Since the chess program IBM Deep Blue defeated world champion Garry Kasparov in 1997, the AIs game became more advanced and managed to beat people in increasingly sophisticated games.
Even in other abstract strategy games like Shogi and Go, both of which are significantly more complex than chess, they are also dominated by machines.
The novelty of AlphaZero, developed by David Silver and his colleagues in DeepMind, is that he avoids the need for information received by people but instead can learn chess, shogovi and play by playing multiple times against each other, one of games
The system managed to beat the most advanced AI programs specializing in these three games after just a few hours of self-training.
As a result, researchers say they should now turn to a new generation of games, such as multiplayer videogames, to cope with the next challenges for AI systems.
<! – Download an attached document from this news ->