Good news for humans: Poker players beat artificial intelligence, for now

There is good news for people who are worried about the robot uprising. Computers might be better at chess but they still lag behind in poker, for now. In a competition that pitted Carnegie Mellon University’s Claudicio program against four of the world’s best poker players, the humans won. They did not, however, win by much.

In a two-week, 80 thousand hand, competition at Rivers Casino the humans ened with a $732,713 collective lead over the program. Unfortunately, with a total of $170 million on the table the size of the victory is not considered “scientifically reliable” or “statistically significant”. The margin, in other words, was thin enough to be considered a statistical tie.

“We knew Claudico was the strongest computer poker program in the world, but we had no idea before this competition how it would fare against four Top 10 poker players. It would have been no shame for Claudico to lose to a set of such talented pros, so even pulling off a statistical tie with them is a tremendous achievement,” said Tuomas Sandholm, the CMU professor of computer science who directed development of Claudico, in a statement.

When all the chips were tallied, Bjorn Li had a total of $529,033, Doug Polk had $213,671, Dong Kimhad $70,049 and Jason Les trailed Claudico by $80,482. All four of the players rank in the top 10 in the world in two-player, No-limit Texas Hold’em.

“We know theoretically that artificial intelligence is going to overtake us one day. But at the end of the day, the most important thing is that the humans remain on top for now,” said Li.

Each player sat for 20,000 hands against Claudico. Although no actual money changed hands during the game, each player will receive appearance fees from a purse of $100,000.

“Thanks to the online stream, the pros had fans rooting for them from all over the world throughout the challenge, in addition to the local players visiting our gaming floor. It’s been very exciting to see this unfold over the last two weeks, and it was a pleasure to partner with Carnegie Mellon University and host these outstanding players,” said Craig Clark, general manager of Rivers Casino.

In 1996 reigning world chess champion Garry Kasparov defeated the IBM supercomputer “Deep Blue”. One year later Kasparov played a second round of chess against the computer and lost.

Poker is a bit different from chess however. Both are games of strategy, but in poker there are unknown variables (not all of the cards are known). It is also more difficult to bluff, or pretend you hold pieces which you don’t, in chess. According to Sandholm it is these unknowns that have made poker one of the great tests of artificial intelligence.

“Beating humans isn’t really our goal; it’s just a milestone along the way. What we want to do is create an artificial intelligence that can help humans negotiate or make decisions in situations where they can’t know all of the facts,” he said.

The computer’s strategy was created using algorithms, rather than human expertise. With only the rules of the game as input the games were run on the Pittsburgh Supercomputing Center’s Blacklight computer. According to the researchers the same approach could be applied to business transactions, cybersecurity, medicine and other areas.

The program is also advancing incredibly rapidly. Sandholm said that, after this match, he and his team already have ideas for improving the algorithms. They also have 80,000 games worth of data on how top players play to use when constructing Claudico’s successor.

Although not specifically worried about poker, in January a number of notable individuals including physicist Stephen Hawking and Space X and Tesla head Elon Musk called for a reigning in of artificial intelligence.

While acknowledging potentially huge benefits from the technology, the signatories expressed concern about possible pitfalls if AI is left unchecked and given too much power over human lives.

“The primitive forms of artificial intelligence we already have, have proved very useful. But I think the development of full artificial intelligence could spell the end of the human race. Once humans develop artificial intelligence it would take off on its own, and re-design itself at an ever increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete and would be superseded,” said Hawking.

Be social, please share!

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmail