Image: Flickr/rafael sergi
When Claudico, a poker-playing algorithm running on a supercomputer, set out to beat four of the world's best no-limit Texas Hold'em players in a competition spanning 80,000 hands two weeks ago, its odds of winning looked slim. Unfortunately for the computer, its chances didn't improve, and it lost to the humans by $732,713 on Friday.That number might seem huge, but considering that the competition—set up by the Carnegie Mellon University researchers who designed Claudico—saw $170 million being bet over the course of two weeks, less than a million dollars is close enough to call it a draw, according to Claudico's creators. The sum lost by Claudico is less than one half of 1 percent of the entire pot."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," said Tuomas Sandholm, one of the researchers behind Claudico, in a statement. "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."No-limit Texas Hold'em, the game the algorithm "played," is an intensely complex one, involving uncertainty and deceit since players can bet big and bluff. There are 10^161 possible choices to be made at any time, and the computer has to decide which to ultimately act on. To handle the computational load, Claudico looked at the "whole game," or all the possible choices, in a simpler, more abstract form. Still, the calculations were complex enough that Claudico needed to crunch data on a supercomputer at the Pittsburgh Supercomputing Center.Even though Claudico lost, it more than held its own against its human counterparts, which might indicate that computers are getting pretty good at dealing with huge amounts of choice and uncertainty. A computer that can make a choice in an ambiguous situation with incomplete information could, for example, one day make a decent lawyer.So, the computer lost this time—and not by much—but just like when you're at the table, things can always change in the next hand.