Artificial intelligence (AI) programs have bested humans in checkers, chess, go and two-player poker, but multiplayer poker was always believed to be a bigger ask. Researchers at Carnegie Mellon University, working with Facebook's AI initiative, on Thursday announced that their program defeated a group of top professionals in six-player no-limit Texas Hold'em. The program, Pluribus, and its big wins were described in the US journal Science. "Pluribus achieved superhuman performance at multiplayer poker, which is a recognized milestone in artificial intelligence and in game theory," Carnegie Mellon computer science professor Tuomas Sandholm said. Sandholm worked with Noam Brown, who is working at Facebook AI while completing his doctorate at the Pittsburgh-based university.
In recent years there have been great strides in artificial intelligence (AI), with games often serving as challenge problems, benchmarks, and milestones for progress. Poker has served for decades as such a challenge problem. Past successes in such benchmarks, including poker, have been limited to two-player games. However, poker in particular is traditionally played with more than two players. Multiplayer games present fundamental additional issues beyond those in two-player games, and multiplayer poker is a recognized AI milestone.
Computer programs have shown superiority over humans in two-player games such as chess, Go, and heads-up, no-limit Texas hold'em poker. However, poker games usually include six players--a much trickier challenge for artificial intelligence than the two-player variant. Brown and Sandholm developed a program, dubbed Pluribus, that learned how to play six-player no-limit Texas hold'em by playing against five copies of itself (see the Perspective by Blair and Saffidine). When pitted against five elite professional poker players, or with five copies of Pluribus playing against one professional, the computer performed significantly better than humans over the course of 10,000 hands of poker. Science, this issue p. 885; see also p. 864
An artificial intelligence called Pluribus has emerged victorious from a marathon 12-day poker session during which it played five human professionals at a time. Over 10,000 hands of no-limit Texas hold'em, the most popular form of the game, Pluribus won a virtual $48,000 (£38,000), beating five elite players who were selected each day from a pool who agreed to take on the program. All of the pros had previously won more than $1m playing the game. What counts as a beating for humanity ranks as a milestone for AI. No computer program has ever achieved superhuman performance against multiple poker players.
The biggest deal in the poker calendar kicks off tomorrow, but only humans will be allowed to sit in on the final table of the World Series of Poker's main event. This Las Vegas knock-out tournament whittled 7,319 entrants down to nine back in July, and after a four-month wait to build media interest in the mainly no-name internet-honed players, this merry band can now get down to playing for the $8.9m (£5.5m) top prize. But has poker bot development reached a level where a bot could be in with a shout for this prize? The short answer is not yet, unless it got lucky. In the short term, luck can override skill, but over time a good player will take a lesser opponent to the cleaners.)