Researchers at the University of Alberta, Canada, think they've made the perfect poker-playing program – and are inviting people to try their hand against it.
The software, dubbed Cepheus, is a machine-learning system that has been taught to play a variant of Texas Hold 'em called heads-up limit, where players can only bet fixed amounts and do so a fixed number of times.
"It was trained against itself, playing the equivalent of more than a billion billion hands of poker," said Michael Bowling, a professor in the university's faculty of science.
"With each hand it improved its play, refining itself closer and closer to the perfect solution. The program was trained for two months using more than 4,000 CPUs each considering over six billion hands every second. This is more poker than has been played by the entire human race."
The results of this research, published in the journal Science, will help improve game theory algorithms so smarter computer systems can be built, and it could be used to aid decision making in software for airport checkpoints, air marshal scheduling and coast guard patrolling, we're told.
"The breakthroughs behind this result are general algorithmic advances that make game-theoretic reasoning in large-scale models of any sort more tractable," said Bowling.
"With real-life decision-making settings almost always involving uncertainty and missing information, algorithmic advances - such as those needed to solve poker - are needed to drive future applications."
The success of Cepheus doesn't mean we're going to see AI win a world series of poker any time soon. Heads-up limit Texas Hold 'em is a fairly basic game with a limited number of steps to take into account and bluffing may cause it a problem too.
You can play the system yourself online, and have a go at beating the machine. If anyone manages it, let us know. ®