Computer to play 2 Poker Pros for 50k
 

Next week, Polaris, a poker-playing computer program built by Schaeffer and his colleagues, will challenge two poker professionals in a $50,000 man versus machine poker game in Vancouver, British Columbia, as part of the annual conference for the Association for the Advancement of Artificial Intelligence.

That's a quote from the article. The article is mainly about a computer that can't lose at checkers. Thought you guys might find it interesting, gonna google for this upcoming match.
LINK TO ARTICLE

Re: Computer to play 2 Poker Pros for 50k
 

Ok found this from Google. a little hard to read because it's translated by Google.
It's Phill Laak and Ali Eslami who are gonna play the computer.

Article

Here is a better article. One sentence says something about limit on betting, so think it's limit.

Article

Re: Computer to play 2 Poker Pros for 50k
 

Big difference if this is limit or NL...

Also, 2000 hands is laughable

Re: Computer to play 2 Poker Pros for 50k
 

Ali Eslami is a high limit limit holdem player who plays mainly at the Commerce.

Re: Computer to play 2 Poker Pros for 50k
 

it has to be limit. i heard they are still working on NL bot

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Big difference if this is limit or NL...

Also, 2000 hands is laughable

[/ QUOTE ]

I think 2000 hands HU is not as laughable as you might think. Usually if you have an edge on someone heads up, they need to run sick good to beat you for 1000+ hands..

If its a ring game, i agree 2000 hands is nothing..

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
I think 2000 hands HU is not as laughable as you might think. Usually if you have an edge on someone heads up, they need to run sick good to beat you for 1000+ hands..

If its a ring game, i agree 2000 hands is nothing..

[/ QUOTE ]

???

I was under the impression that 2K hands has better implications for long term results @ full ring than HU?

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
[ QUOTE ]
Big difference if this is limit or NL...

Also, 2000 hands is laughable

[/ QUOTE ]

I think 2000 hands HU is not as laughable as you might think. Usually if you have an edge on someone heads up, they need to run sick good to beat you for 1000+ hands..

If its a ring game, i agree 2000 hands is nothing..

[/ QUOTE ]

You are semi-correct. HU against bad opponents has less variance. HU between competent opponents probably has the most variance of any game imaginable.

But 2k hands in any form of poker is going to be high variance simply because it isn't enough hands to evenly distribute similar strength hands in equal proportions (as in, Comp might get AA 15 times, while pro gets them 6). It's like flipping a coin 10 times.

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Big difference if this is limit or NL...

Also, 2000 hands is laughable

[/ QUOTE ]

I think 2000 hands HU is not as laughable as you might think. Usually if you have an edge on someone heads up, they need to run sick good to beat you for 1000+ hands..

If its a ring game, i agree 2000 hands is nothing..

[/ QUOTE ]

You are semi-correct. HU against bad opponents has less variance. HU between competent opponents probably has the most variance of any game imaginable.

But 2k hands in any form of poker is going to be high variance simply because it isn't enough hands to evenly distribute similar strength hands in equal proportions (as in, Comp might get AA 15 times, while pro gets them 6). It's like flipping a coin 10 times.

[/ QUOTE ]

Actually that makes a ton of sense, especially if the computer is essentially playing not to lose instead of playing to win. In which case it would be literal coin flips, because it would be pretty damn close to unexploitable. But you are right the smaller the edge, means much bigger variance, and probably especially headsup.


I stand corrected.

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Big difference if this is limit or NL...

Also, 2000 hands is laughable

[/ QUOTE ]

If this was vs a computer couldnt they have got a decent online player to 8 table it so they could get a number of hands that were at least semi statistically significant, 2k is such a joke

Re: Computer to play 2 Poker Pros for 50k
 

i have a lava lamp that could beat phil laak at limit poker.

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
i have a lava lamp that could beat phil laak at limit poker.

[/ QUOTE ]

Only online, the proper players have such a hudge advantage over electronic lighting devices live

Re: Computer to play 2 Poker Pros for 50k
 

2k hands is good for the pros IMO since it gives the computers a lot less time to create an accurate model.

Re: Computer to play 2 Poker Pros for 50k
 

Live bots usaully have big tells like blinking lamps when they are going to checkraise, and they often scream out DANGER DANGER when they are bluffing

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Live bots usaully have big tells like blinking lamps when they are going to checkraise, and they often scream out DANGER DANGER when they are bluffing

[/ QUOTE ]

LOL, very good.

Re: Computer to play 2 Poker Pros for 50k
 

MasterLJ,
four 500 hand !duplicate! matches = 4000 hands. This may yield a statistically significant result.

Popinjay,
the program does no opponent modeling, it is designed to play in a gametheoretically optimal way.

Re: Computer to play 2 Poker Pros for 50k
 

this is retarded cause one time heads up matches don't mean [censored]. they need to make this a [censored] series or something

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Big difference if this is limit or NL...

Also, 2000 hands is laughable

[/ QUOTE ]

Though the hands are randomly generated, they will be mirror matches, which is why they are using two pros.

In hand #354, it might be AJo vs. 44 on a board of AJ4T9. One of the human players will have AJo playing against the program's 44 and the other human player will have 44 playing against the program's AJo.

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Live bots usaully have big tells like blinking lamps when they are going to checkraise, and they often scream out DANGER DANGER when they are bluffing

[/ QUOTE ]

wp

Re: Computer to play 2 Poker Pros for 50k
 

phil laak any good?

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Ali Eslami is a high limit limit holdem player who plays mainly at the Commerce.

[/ QUOTE ]

Yeah, and limit hold'em isn't his best game at all IMO, though he is certainly tough, but he's not like Joe Cassidy tough. I wonder how he got affiliated with this, he isn't really known by the casual poker fan.

-DeathDonkey

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Live bots usaully have big tells like blinking lamps when they are going to checkraise, and they often scream out DANGER DANGER when they are bluffing

[/ QUOTE ]

ahjaahahahahahha

Re: Computer to play 2 Poker Pros for 50k
 

they should let me play the bots. i can crush pokeracademy bots all day.

Re: Computer to play 2 Poker Pros for 50k
 

Re: Computer to play 2 Poker Pros for 50k
 

have these bots ever played any good human players before??

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
[ QUOTE ]
Big difference if this is limit or NL...

Also, 2000 hands is laughable

[/ QUOTE ]

Though the hands are randomly generated, they will be mirror matches, which is why they are using two pros.

In hand #354, it might be AJo vs. 44 on a board of AJ4T9. One of the human players will have AJo playing against the program's 44 and the other human player will have 44 playing against the program's AJo.

[/ QUOTE ]

that's pretty awesome

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
phil laak any good?

[/ QUOTE ]

Jen Tilly says he very good and gives her multiple . . .
Oh wait, no idea how he is at limit poker.

Re: Computer to play 2 Poker Pros for 50k
 

Wouldn't online players have a monster advantage vrs a computer.. and arent they better to begin with .. makes ense?

Re: Computer to play 2 Poker Pros for 50k
 

Laak is good enough to beat up on some HSNL live games (like really big) but as I understand from reading some of his stuff he has awsome game selection (not to say that he is very good and capable, I assume he is)

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
Laak is good enough to beat up on some HSNL live games (like really big) but as I understand from reading some of his stuff he has awsome game selection (not to say that he is very good and capable, I assume he is)

[/ QUOTE ]

Well it doesn't really matter if either pro is the best, over even that great compared to other high stakes pros. if the computer plays at all competitively, which i am almost certain it will, then that is pretty strong evidence of computers being capable of earning a programmer a living at HU poker.

also, just a general LOL to all the people who think poker is so special that computers won't be able to compete at it. Humans are horribly are estimating statistics, and behaving randomly two things that are very important in poker.

also short and medium stack NL holdem would be much easier to solve than limit holdem, far fewer decisions.

Re: Computer to play 2 Poker Pros for 50k
 

IIRC Laak has some sort of an engineering background which may give him more insight into the weaknesses of the computer than someone who's just an uneducated good poker player.

Re: Computer to play 2 Poker Pros for 50k
 

While its obvious that a 2000 hand match isn't going to definitively prove anything, the fact that the computer owner is willing to put up serious cash says a lot about the state of poker AI.

Re: Computer to play 2 Poker Pros for 50k
 

Phil Laak is pretty damn good Poker player overall, Limit, NL, PL.

That said, I think he loses a lot of his ede when his opponent is inanimate and can't/won't respond to his "routine."

Re: Computer to play 2 Poker Pros for 50k
 

[ QUOTE ]
four 500 hand !duplicate! matches = 4000 hands. This may yield a statistically significant result.

[/ QUOTE ]

The "duplicate" thing will sure help but it has much to do with the styles too; if the opponent is pushing small edges by heavy aggression or by roping down a lot, then the fluctuations when one player only has a marginal edge (like 1.5 big bets per 100 hands) can give results like five big bets won per 100 hands during the first 5k of hands and five big bets lost per 100 hands during the next 5k hands. Then there are players who do a lot of e.g. (check-)raising rather than roping and the opponent plays it solid; in that combined case the fluctuations are much smaller, one being a heavy favorite to be ahead after 5000 hands even with a small edge.

If the computer or the humans get beaten in a big way like five big bets per 100 hands, it could be because of the fluctuations, but in case the humans say it was an easy game that they really did have that much of an edge, then there's about no way the computer is going to do anything but lose heavily. What those players will say about it is more important than the fluctuations, no matter what the result. As far as my experience against bots goes, they have been at least as bad as weak-passive real life opponents, and that's more true the less players on the table, e.g. at full ring they can do pretty well but as soon as it gets shorthanded they start losing much more, and heads up they get beaten even more.

One edge that the computer can have over humans is that humans have experience against humans, not against bots, and the match is short enough that there's not all that much time to learn, though during 2000 hands one can learn a lot, though the computer is made unpredictable enough so it will have some edge for a start, except that both players do rate to have experience against bots as they say themselves, and only if the bot will be very different the humans in the future may start to ask to practise against it before the match, like they do at chess, though in that chess case the computer can study their opponent's style before the match while that may not be so much a factor in this case.

I am not too happy about the factor that the computer is not adjusting its optimal strategy during the match, though it's also possible to use that adjustment against the computer - there's no problem in the real life as the opponents do not know they are playing against a bot. Then I also believe that if the computer is using the right style, it can't be exploited much even when one gets a complete line, but it would be better if the computer would vary its style during the game more or less randomly, though there is something like that in its strategy, so it's not playing just the same "optimal" strategy all the time as far as I understood. Additionally it could trick humans that when they start to adjust it picks a strategy that does well against that adjustment, and the humans will not know that.

Re: Computer to play 2 Poker Pros for 50k
 

If the computer only needs to break even to get the $50K, I think I could program it to achieve that, given the "mirror" conditions.