Cepheus, programmed by researchers at The University of Alberta has a strategy for two-player limit Texas hold'em poker that is unbeatable by a human poker player.

According to team leader Michael Bowling:

"Poker has been a challenge problem for artificial intelligence going back over 40 years, and until now, heads-up limit Texas hold'em poker was unsolved"

Unlike chess, where the pieces are visible to both players, poker is a games characterised by imperfect information, where players do not have full knowledge of past events. Texas hold'em is the most popular variant of the game and when played with just two-players (heads-up) and with fixed bet-sizes and number of raises (limit), it is called heads-up limit hold'em.

The imperfect information nature of heads-up limit hold'em made it a challenging game for computers to play or solve but now that milestone has been achieved.

In the offical announcement from the University of Alberta, Bowling says:

"We define a game to be essentially solved if a lifetime of play is unable to statistically differentiate it from being solved at 95% confidence. Imagine someone playing 200 hands of poker an hour for 12 hours a day without missing a day for 70 years. Furthermore imagine them employing the worst-case, maximally exploitive, opponent strategy, and never making a mistake."  

The breakthrough was reported on January 9 in Science in an article that explains how the result was enabled by a new algorithm, CFR⁺, which is capable of solving extensive-form games orders of magnitude larger than previously possible and also reveals that the researchers' computation formally proves the common wisdom that the dealer in the game holds a substantial advantage. 

You can find out more, and try playing against it, on the Cepheus website. 

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info