research shop owned by Google-parent Alphabet Inc., beat Lee Sedol, the world’s best player at the game. In 2016, AlphaGo, an algorithm created by DeepMind, the London-based A.I. It, of course, succumbed to artificial intelligence in 1997 when IBM’s DeepBlue algorithm beat grandmaster Gary Kasparov. Chess was long considered the epitome of human strategic thought, a symbol of calculating rationality and intellect. In order for such software to judge whether a particular action is likely to be beneficial, points serve as a convenient reward signal, in much the way a dog trainer doles out a treat if Fido sits on command. This makes them ideal environments for reinforcement learning, a technique where software learns from experience instead of existing data. Computer scientists have also favored games for another reason: they have point systems and clearly-defined winners and losers. Games test reasoning ability and simulate, in simplified form, some of the decision-making dilemmas found in the real-world. Games have long-been used as benchmarks of A.I. He adds, however, that neither he nor Facebook have any immediate intention of commercializing the technology. Another area, he says, is cybersecurity-where adversaries have imperfect information about one another’s capabilities and intentions. Business negotiations are one area where the techniques behind the Pluribus might be useful, Brown says. “This could be deployed across the board to countless scenarios,” he says. Brown is a researcher at Carnegie Mellon and also holds a position at Facebook’s A.I. “Most real-world strategic interactions involve hidden information or multiple opponents or both,” Noam Brown, one of the researchers who designed Pluribus, says. The development has potential implications for the business world where, as in poker, people need to make strategic choices amid great uncertainty and where outcomes are often not zero-sum. The research was jointly undertaken by Facebook and Carnegie Mellon University in Pittsburgh, Pennsylvania. The feat, which was announced today in conjunction with the publication of a research paper on the experiment in the journal Science, marks an important milestone in the development of artificial intelligence. Among the players the bot, which is called Pluribus, beat were four-time World Poker Tour champion Darren Elias as well as World Series of Poker Main Event champions Chris “Jesus” Ferguson and Greg Merson. learned entirely by playing millions of hands against itself, with no guidance from human card sharks. A poker bot has beaten a table full of pros at six-player, no-limit Texas Hold ’em, the version of the game used by most tournaments, over the course of 10,000 hands of play. And, unlike in the old Kenny Rodgers ballad, it didn’t need a grizzled cowboy gambler to teach it a trick or two. It knows when to hold ’em and when to fold ’em.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |