![]() I'd like to report that it all came back to me, that some long dormant part of my brain where I kept the skills from my one of my favorite childhood board games sprung to life on the battlefield. McCaffrey concluded, "Either way, the Stratego program is an interesting new development in deep reinforcement learning.I won the first game of Stratego I played in probably 15 years. Can the fundamental ideas be applied to useful applications, or are such systems strictly one-off applications that don't readily generalize?" McCaffrey added, "The key question however, is the extent to which the ideas used programs like DeepNash can generalize. ![]() McCaffrey commented, "The DeepNash program is impressive and represents a significant step forward on the path to mastering complex strategy games with imperfect information." James McCaffrey from Microsoft Research, who has experience with AI approaches to games, and experience with deep reinforcement learning, the basis of DeepNash. The opponent will use his Marshal (10) to chase the piece thought to be a General and be lured to pass the Spy waiting in ambush. Then a moderate-strength piece can chase a high-value exposed opponent piece such as a Colonel (8), bluffing that the chasing piece is a General (9). The Spy can be placed in a central location and wait. For example, the only piece that can defeat a Marshal (10) is the lowly Spy (1). ![]() The DeepNash program learned a variety of bluffing tactics. This includes using initial piece placements that vary enough to prevent its opponent from spotting patterns over a series of games. To become difficult to exploit, DeepNash develops unpredictable strategies. The research abstract states that DeepNash uses a game-theoretic, model-free deep reinforcement learning method called the Regularized Nash Dynamics (R-NaD) algorithm.Īs of 2022, the DeepNash program beats existing state-of-the-art AI methods in Stratego and achieved an all-time top-three rank on the Gravon games platform, scoring a win rate of 84 percent against human expert players. The name is a reference to deep reinforcement learning and mathematician John Nash (1928-2015). The new Stratego computer program is called DeepNash. The key to success in Stratego is information extraction - determining the types of the opponent's pieces. Three of the Blue pieces have been exposed by defeating a Red piece. The image in Figure 3 shows a typical game in progress from the point of view of the Red player. Games between experienced human players typically last about 300 moves and take anywhere from about 20 minutes to two hours. The Flag piece is almost always placed on the first row where it's less vulnerable, and the Flag is usually surrounded by two or three of the Bomb pieces for protection. But if the Marshal or any other piece lands on the Spy, the Spy loses. The Spy piece, with value = 1, can land on and ambush the Marshal that has value = 10. It usually sacrifices itself to identify the type of an opponent's piece. The special Scout piece, with value = 2, can move multiple squares in a straight line like a chess Rook. If both pieces have the same value, both are removed. When a piece lands on an opponent's piece, the piece with the lower point value loses and is removed. The regular pieces move one square at a time to an adjacent square (not diagonally). The goal of the game is to capture the opponent's Flag. Each player, Red and Blue, has 40 pieces, one of which is the Flag. There are two 2x2 blocks at the center of the board where pieces cannot move. The image in Figure 1 shows an example starting position. In order to understand the new approach to playing Stratego, it's necessary to have a basic understanding of the rules of the game. The two key challenges for a software program are that the game has imperfect information (opponent pieces are hidden, which allows bluffing) and sheer size - the number of possible positions vastly exceeds the number of atoms in the known universe. However, Stratego is extraordinarily difficult for computers. The strategy board game known as Stratego (the name is copyrighted, but there are many versions of the same game with different names) is deceptively simple and can be played by children. Other games that have been successfully tackled to a large extent include poker, Go, video games and Diplomacy. Chess was essentially conquered when the Deep Blue program beat world champion Garry Kasparov in 1997. Researchers have explored applying artificial intelligence to game playing for decades, starting with chess in the 1950s. The technique is explained in the 2022 research paper "Mastering the Game of Stratego with Model-Free Multiagent Reinforcement Learning" by J. Researchers at Google have demonstrated a new algorithm that achieves superhuman performance on the Stratego board game.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |