Jonathan Schaeffer Answers 2 Questions About Polaris the Poker Bot


Yesterday I wrote a post about the upcoming Texas Hold’em match (July 23 and 24) between Polaris, the  poker-playing computer programmed by the computer science team at University of Alberta, and two humans, Phil Laak and Ali Eslami.  The post received a question from Sparky about where the name ‘Polaris’ comes from.  I couldn’t find an answer to this question by searching on the Web, so I decided to go straight to Jonathan Schaeffer and ask him for the answer.

Professor Jonathan Schaeffer holds the Canada Research Chair in Artificial Intelligence and is the team leader of the Polaris program.  He was kind enough to give me an answer to Sparky’s question.  He writes,

There is nothing special in the name, other than it reflects that we are the “northern star” in the computer poker world.  It is a subtle hint that we are Canadian 🙂

Of course, I couldn’t help myself and just ask one question.  I explained to Professor Schaeffer that I am interested in the concept of luck as it applies to Texas Hold’em, and I asked him whether his Polaris experiment would eliminate all luck as a factor.  And if it didn’t, then in what way did he think that luck would still play a factor in the match?  Here’s his answer.

It will eliminate most of the luck factor, but not all of it. That is why there is a rule in place for a drawn match. If one side wins by less than 25 small bets (500 hand match) then the match is declared a draw because the result is statistically insignificant.

(Thanks for your answers, Jonathan, and I look forward to seeing your silicon Bot kick some carbon butt! – Suber)

Leave a Reply

Your email address will not be published. Required fields are marked *