I have a son-in-law who plays a mean 'Texas Hold 'em.' (That's a sophisticated poker game for you high-stake
wannabes who have 'TELL' written all over your bodies.)
|Not Poker Face|
I don't like to see the other players sweat and squirm...
who wouldn't like that?... but because the answer to the unbeatable perfect-poker-playing probot (think Peter Piper... ) could be the key to saving the world. Really--sort of. It's a 'greater good' kind of thing.
This could be THE break-through to better solutions in so many human applications. (For you 'robotophobes,' don't worry, this isn't artificial intelligence that thinks--YET... but don't hold your breath.)
This is a highly sophisticated algorithmic solution that does more than play unbeatable chess, perfect Scrabble and a mean Jeopardy. This bugger plays poker where so much information about state of play is hidden. The poker-playin' program should be called Tex... a solid poker-player name instead of some nerdy-sounding thing like Cephus (which it is now).
"Our goal," says Dr. Bowling of Canada's University of Alberta, "is to advance artificial intelligence. And poker is an ideal game to capture all kinds of uncertainty."
To learn all the nuances of bluff, bet, hole cards and luck of the draw, the algorithm ran on an array of 4,000 computers for 68 days, calculating the optimum outcomes for more than a trillion possibilities of play.
The program worked through six billion hands every second, more than ever played by all the people who have ever been on earth, and it refined itself to near perfection. If you played against it, no matter what you do, you still wouldn't be able to eke out any more than a tiny advantage over millions and millions of hands."
But, as any true gambler knows, "This could be my lucky day."
Researchers don't plan to commercialize the system but do plan on setting up a University-managed public website where you can try your skill against it for free.