The 20-day poker tournament between four human pros and an artificial intelligence program concluded last night. The AI, named Libratus, was created by researchers at Carnegie Mellon University and bested its opponents by $1.76 million in chips. “The best AI’s ability to do strategic reasoning with imperfect information has now surpassed that of the best humans,” said Libratus co-creator Tuomas Sandholm in a press release.
While the AI has so far only been applied to poker, CMU is eager to experiment with its abilities in other fields. “The computer can’t win at poker if it can’t bluff,” said Frank Pfenning, head of the Computer Science Department at CMU. “Developing an AI that can do that successfully is a tremendous step forward scientifically and has numerous applications. Imagine that your smartphone will someday be able to negotiate the best price on a new car for you. That’s just the beginning.”
There was intense speculation during the tournament about how Libratus was able to shift its strategy so dramatically from day to day. A number of poker pros weighed in to say that the researchers must be tinkering with it by hand at night. But the CMU team shot down that speculation this morning. In a press release they described the process:
The event was surrounded by speculation about how Libratus was able to improve day to day during the competition. It turns out it was the pros themselves who taught Libratus about its weaknesses. “After play ended each day, a meta-algorithm analyzed what holes the pros had identified and exploited in Libratus’ strategy,” Sandholm said. “It then prioritized the holes and algorithmically patched the top three using the supercomputer each night. This is very different than how learning has been used in the past in poker. Typically researchers develop algorithms that try to exploit the opponent’s weaknesses. In contrast, here the daily improvement is about algorithmically fixing holes in our own strategy.”
Libratus relied on the Pittsburgh Supercomputing Center’s Bridges computer, which, according to CMU, has a “total speed is 1.35 petaflops, about 7,250 times as fast as a high-end laptop and its memory is 274 Terabytes, about 17,500 as much as you’d get in that laptop.”
Most online poker players have nothing to fear from Libratus right now. The system only works in Heads Up poker, where only two players are involved. A game with three players or more would be too computationally intensive, and require a totally different strategy and algorithmic approach. And the human players from this tournament were able to find a silver lining. For 20 days of work, they each won a share of a $200,000 prize pool, based on their performance, despite losing big to Libratus. “Usually, you have to lose a lot and pay a lot of money for the experience,” said Jason Les, one of the pros, in today’s press release. “Here, at least I’m not losing any money.”