JazzyBets

Over on the minimalist gambling news blog, Smash-Vegas, there is an excellent series of beginner articles on the “Anatomy of a Pokerbot” , using the open source pokerbot software, OpenHoldem. The series of posts on Smash-Vegas covers a wide range of PokerCoder challenges when developing a Pokerbot from scratch, such as the importance of stealth, improving the AI development with Pokertracker software, and using screen scrapers effectively.

The collection of posts of Smash-Vegas also covers introduction to Pokerbot strategy, and a brief, informative riff on First steps in Coding a Pokerbot. An interesting resource revealed in the pokerbot post is the OpenHoldem End-User documentation and formulas.

Taking a last look at the Man vs. Machine Pokerbot championship last month, the University of Alberta research team leading had an amusing quote that “real world problems are more like poker than chess” after the match on MSNBC.

“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.”

Some the biggest winning players in online poker history are gathering together this summer for a match-up with the infamous Polaris 2- the most technologically sophisticated poker Artificial Intelligence (AI) machine in the world. This event is the second edition of the Man vs. Machine Poker Championship and is scheduled to take place during the 2008 Gaming Life Expo at the Rio All-Suite Hotel & Casino in Las Vegas July 3-6.

Polaris 2 was created by the University of Alberta’s Computer Poker Research Group (http://poker.cs.ualberta.ca) and is the result of several years of scientific research and thousands of man-hours of development of the Pokerbot logic.

The human players slated to join in the match are professional poker coaches or contributors to the widely recognized online poker training site Stoxpoker.com. Each of the human players participating in the challenge has won millions playing for the highest stakes against some of the best poker players in the world.

Bryce Paradis, head coach at Stoxpoker.com and winner of over 2.5 million dollars at online poker in 2007, added that “Against the current AI in Polaris 2, the average poker player would be completely dominated. The Polaris 2 team has made incredible improvements since the match last year. The most powerful change is that the AI will now learn from and adapt to its opponents’ play as the match progresses. This year’s Man vs. Machine match is going to push our team to their limit.”

In 2007, Polaris 1 competed with professionals Phil “The Unabomber” Laak and Ali Eslami in the city of Vancouver, Canada. The series ended in a very close finish, with a final score of two wins, one loss, and one statistical tie for the human team of Phil and Ali. Each round comprised of a 500 duplicate hand match, wherein the same series of cards was dealt to both players, with each one playing opposite hands in the game with the AI taking the other side. At the end of the match the total number of chips won or lost by each team determined the winner. In an effort to reduce the element of random luck to a minimum, the same format will be used again this year.

The key members of the Stoxpoker.com team are professionals Nick Grudzien, IJay Palansky, as well as Stoxpoker contributor Matt Hawrilenko, all of whom can boast well over $1M in lifetime cash-game limit hold’em winnings. Additional professional players are expected to round out the team in the coming weeks.