In 2017, a poker bot named "Libratus" developed by researchers
at Carnegie Melon University (CMU) led by Professor Tuomas Sandholm
and Ph.D. student Noam Brown, beat some of the best heads-up poker
pros in the world in Texas hold'em over a large sample size. The
advancement was considered a milestone at the time, but its
applications were limited due to the binary task of beating just
one opponent at a time — in heads-up play.
The latest poker bot developed by the same researchers in a joint project between Facebook AI and CMU was able to do something that no other AI has achieved — beat multiple strong players in the incomplete information game of no-limit hold'em in a six-handed format, and it did so more efficiently than any other documented poker bot before it.
Speaking with Kennedy, four-time World Poker Tour champion Darren Elias explains that he helped train Pluribus by competing against four tables of bot rivals and alerting scientists when the A.I. made a mistake. Soon, the bot “was improving very rapidly, [going] from being a mediocre player to basically a world-class-level poker player in a matter of days and weeks.” The experience, Elias says, was “pretty scary.”
According to the Verge’s James Vincent, Pluribus—a surprisingly low-cost A.I. trained with less than $150 worth of cloud computing resources—further mastered poker strategy by playing against copies of itself and learning through trial and error. As Jennifer Ouellette notes for Ars Technica, the bot quickly realized…