DeepMind’s AI DeepNash has reached the level of professional human players in Stratego. Stratego is a board game that requires strategic thinking in the absence of information. It is much more complex than chess, go or poker that AI has mastered before. In the game, each side has 40 pieces, and each other cannot see the true “identity” of the pieces. The two sides take turns to move the chess pieces to eliminate the opponent’s chess pieces, and the party that wins the opponent’s flag or destroys all the movable chess pieces finally wins. Therefore, players need to make strategic deployments, collect information, and play games with each other. There are 10^535 possible layouts in a game of Stratego. In contrast, there are 10^360 possible layouts in Go. Furthermore, in Stratego, the AI needs to reason about more than 10^66 deployment strategies for the opponent, which dwarfs the initial 10^6 possible situations in Texas Hold’em. DeepNash is named after the mathematician John Nash who proposed the Nash equilibrium. DeepNash combines reinforcement learning algorithms with deep neural networks to find Nash equilibria.
This article is transferred from: https://www.solidot.org/story?sid=73583
This site is only for collection, and the copyright belongs to the original author.