— One year after a famous man-vs.-machine poker tournament, the machine finally won out over a team of living, breathing poker professionals. The University of Alberta's Polaris poker-playing software came from behind for the victory in a six-round match held July 3-6 at the Rio All-Suite Hotel and Casino in Las Vegas.
The outcome wasn't clear until the final scores were tallied up from the final game. Each of the games involved 500 hands of limit hold 'em, with Polaris playing head-to-head against each of the professionals. In the end, Polaris won three of the games, the humans won two, and one was a draw.
Polaris' programmers rejoiced.
"It's hard to describe how good that felt," research team leader Michael Bowling said in a report from the University of Alberta. "As a group, we may not all be great poker players, but all of us really, really want to win."
Bowling pointed out that this was just a first step. "This was really the simplest form of poker," he said. "There's a lot more we can look at, such as playing without betting limits, or playing with more than two opponents. One of the reasons I got excited about this line of research is that it's not just a one-off. It's a really challenging path of research."
Historically, computers have been better at games where all the information is essentially out there on the board - for instance, chess and checkers. Poker is trickier, because players have to make judgments based on different amounts of information about the state of play.
"In general, problems in the real world are going to be more like poker than chess," Bowling said.
Polaris and other pokerbots have been around for years, and it's almost a given that some of those bots have been employed on online gaming sites against not-so-professional human players. Check out this archived report on the subject from Michael Brunker, one of my colleagues at msnbc.com, and this more recent report on a real-money pokerbot (it's actually the first part of an ongoing series).
If robo-poker isn't your cup of tea, here are some other weekend field trips you can take on the Web:
- The nonprofit Planetary Society says it is extending the deadline for sending your name to the moon. Now you have until July 25 to add your name to the list. The names will be digitized, stored on a microchip and placed on NASA's Lunar Reconnaissance Orbiter, which is due for launch late this year. As I noted back in May, you can even print out a certificate recognizing your participation in the project.
- Something funny seems to be going on at the sun. Or more accurately, not going on. The sun is at the low point of its 11-year activity cycle, and it's been that way for going on three years. Some observers have wondered whether the solar minimum is lasting an abnormally long time, but on the Science @ NASA Web site, solar physicist David Hathaway says the current quiet phase "is well within historic norms for the solar cycle." Space.com had a story last month about the eerie calm and is passing along the calming follow-up today.
- So how are we doing? A survey of more than 1,300 scientists from around the world, published in this week's issue of the journal Science, indicates that scientists now think science writers aren't so bad after all. "Scientists actually see rewards in this process, not just pitfalls," study co-author Sharon Dunwoody, a journalism professor at the University of Wisconsin at Madison, said in a news release about the survey. The respondents included scientists from the United States, Britain, France, Germany and Japan. You can get the international perspective from University College London and the Helmholtz Association of German Research Centers - and there's even an MP3 audio interview with study director Hans Peter Peters.
The last word, as always, is yours. What do you think about the rise of the pokerbots, the state of the solar cycle or the state of science in society? Feel free to add your comments below.
Update for 4:54 p.m. ET July 13: I've revised the reference to how pokerbots are being used on gaming Web sites, in response to comments below.