This site may earn affiliate commissions from the links on this page. Terms of use.

473191-bad-poker-hand

When was the final time your computer deliberately deceived you? Several of my less-figurer-literate relations would reply "all the time." But this speaks more to our ability to deceive ourselves than it does to any technological intent. That could all change come January xi when a Carnegie Mellon poker-playing AI will face off confronting some of the best poker minds at the Rivers Casino in Pittsburgh, PA.

Information technology's become most banal to speak of automobile intelligence dethroning another board game hero, so what makes this poker tournament special? A successful poker player, be it AI or human being, must possess something akin to a theory of their opponent's listen and the concomitant ability to mislead and bluff — qualities that are notoriously difficult to program into computers. And while we may call up of lying as representing one of the worst features of human nature, with a poker bluff it could likewise exist considered a hallmark of intelligence. To bluff, we must accept an thought well-nigh someone else's expectations, nearly what they anticipate happening in the time to come. Most animals accept a limited power to predict the time to come nether weather of imperfect information, much less develop strategies based upon some other creature's predictions.

brains_vs_ai_853x480-min

Poker Pro Dong Kim facing off confronting an before version of CMU poker AI. Prototype Courtesy of Carnegie Mellon University

You lot might even argue this raised humans to the top of nature's pecking gild: Nosotros have a ameliorate ability to simulate the expectations of other animals and make predictions based upon on those theories. A human being can look at fox tracks in the woods and conclude that a trick, beingness a creature of habit, is probable to come this way again — and if we place a trap in said spot, the fob volition stumble upon information technology. For that reason, building machines that can all-time us in games of imperfect and misleading information might simply be a stride likewise far where our ain planetary dominance is concerned.

We have already seen some crucial advances in this regard, specifically the AlphaGo supercomputer that defeated Lee Sedol at the lath game Become. Unlike poker however, Become is a game of perfect information, meaning both players are aware of their opponent's options and both share the same information gear up. But what makes Become more difficult than chess for a estimator is complexity — numerical estimates show that the number of possible games of Get in exceeds the number of atoms in the observable universe. This means a computer cannot mathematically brute force its fashion into victory and must develop something alike to intuition when developing a strategy. Compare this with the kind of decision-making a human faces when because whether to go out a task in favor of a new 1. In that location is no mode to ensure one is making the best determination under such weather, as there are just too many variables – commute time to the new office, personality chemical science with the new dominate, changing marketplace conditions, and so on. While both Go and poker share this characteristic of enormous complexity, Poker also incorporates features like bluffing, slow playing, and other strategic gambits.

What's scary is that a figurer algorithm, which tin can best humans in a game involving bluffing, imperfect information, and farthermost complication, can be tailored to defeat the states at many of the "games" we play with each other. Think of the bargaining that goes on during a hostile takeover betwixt companies, or during a divorce settlement, or betwixt two parties making a existent estate deal. For that reason, during adjacent week's collision between the poker masters as the motorcar, much more than the $20,000 purse may hang in the balance. I, for one, volition be post-obit the results closely.