sciencetech-blog

comments

It could be bad new for casino's around the world - scientists claim to have created an unbeatable poker playing robot.

The program considered 24 trillion simulated poker hands per second for two months, probably playing more poker than all humanity has ever experienced, says Michael Bowling, who led the project. 

Although robots have beaten human players before, the Canadian team say they have now essentially 'solved' the game.

The team boast their robot is unbeatable by someone playing 200 hands of poker an hour for 12 hours a day without missing a day for 70 years.

The resulting strategy still won't win every game because of bad luck in the cards. 

But over the long run — thousands of games — it won't lose money. 

'We can go against the best (players) in the world and the humans are going to be the ones that lose money,' said Bowling, of the University of Alberta in Edmonton, Canada. 

'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,'

'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,' explains Bowling. 

'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 kind of poker it applies to has waned in popularity over the past seven years or so.

Even online, the stakes tend to be small and 'you'd be winning a few dollars, not raking in millions.'

Bowling said the computer's strategy is far too complicated for anybody to memorize, with about 1,000 times the amount of information in the English-language Wikipedia. 

But his university has created a website where people can ask it for advice and even play against it. 

Poker is a family of games that exhibit imperfect information, where players do not have full knowledge of past events.

The most popular variant of poker today is Texas hold'em. 

When it is 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. 

While smaller than checkers, the imperfect information nature of heads-up limit hold'em makes it a far more challenging game for computers to play or solve.

PLAYING COMPUTER GAMES 

For over a half-century, games have been test beds for new ideas in Artificial Intelligence (AI) and the resulting successes have marked significant milestones - Deep Blue defeated Kasparov in chess and Watson defeated Jennings and Rutter on Jeopardy! 

While many perfect information games (where all players are informed of everything that has occurred in the game prior to making a decision) have been solved, e.g., Connect Four, no nontrivial imperfect information game played competitively by humans has previously been solved. 

These games are more challenging, with theory, computational algorithms, and instances of solved games lagging behind results in the perfect information setting. 

And, while perfect information may be a common property of parlour games, it is far less common in real-world decision making settings.

While many perfect information games (where all players are informed of everything that has occurred in the game prior to making a decision) have been solved, e.g., Connect Four, no nontrivial imperfect information game played competitively by humans has previously been solved. 

These games are more challenging, with theory, computational algorithms, and instances of solved games lagging behind results in the perfect information setting. 

And, while perfect information may be a common property of parlour games, it is far less common in real-world decision making settings.

'The breakthroughs behind this result are general algorithmic advances that make game-theoretic reasoning in large-scale models of any sort more tractable,' says Bowling.

And, while seemingly playful, game theory has always been envisioned to have serious implications, including a surge in game-theoretic applications involving security, such as systems being deployed for airport checkpoints, air marshall scheduling, and coast guard patrolling. 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.