A group of scientists from DeepMind has developed a computer program called Artificial Intelligence (AI), which learns to beat people in the three most sophisticated games thanks to an algorithm that competes with no instructions.
Researchers presented AlphaZero, a computer program capable of not only achieving super-human mastery in some of the most sophisticated board games such as Chess, Shogi and Go but also learning to play without knowledge. before, except for the rules of each game. "
The system was created by experts from DeepMind, a British company owned by Alphabet, the parent company of Google.
The results, published in the journal Science, represent an important step in the development of an AI player who can quickly master and master each game.
AlphaZero's novelty is that it avoids the need for information from people, and instead is able to learn chess, shogi and Go by playing repeatedly against each other while learning each of the games.
From scratch (hence its name), without help or examples of real players. In other cases, the computer learns because it feeds it with all the human knowledge, millions of plays and real examples, and from there the silicon brain chooses the best strategies.
This new system managed to beat the most advanced AI programs specializing in these three games after just a few hours of self-training.
Researchers say they should now turn to a new generation of games, such as multiplayer video games, to cope with the next challenges for the AI systems.