Spooky accuracy in predicting a gambler's next bet

Artificial intelligence works in part to replicate the human mind, and scientists need to know how humans make their next moves, reports Scientific American this week. The article goes on to report that scientists have created a system that can predict the bets of gamblers, with an accuracy “bordering on spooky.”

The research to which Scientific American refers was published July 11th this year by Victor K.Y. Chan in the Journal of Gambling Studies. Chan is a researcher at the Macau Polytechnic Institute, which funded his research.

Chan’s team used data from a total of 675 online games from six Texas Hold ’em gamblers to build a mathematically simulated neural network based on the players’ initial few plays to establish that the network learned and rewired itself based on guesses that either turn out to be right or wrong. This method of AI is called propagation of error.

The researchers found that their neural model could predict each of the six gamblers’ bet amounts with an accuracy to three decimal places of the dollar, and could also predict with similar accuracy their cumulative wins/losses.

The conclusion was that, based on their first few games, the gamblers’ subsequent behavior, strategy and ultimately their wins and losses, was consistently predictable.

Chans findings no doubt add to the body of knowledge accumulated by AI poker studies at Canadian universities. The full and very detailed scientific paper on the Macau achievement can be read here:

http://www.scribd.com/doc/17472377/Using-Neural-Networks-to-Model-the-Behavior-and-Decisions-of-Gamblers-in-Particular-CyberGamblers