backtop

Print E-mail del.icio.us 20 comment(s) - last by oTAL.. on Jul 27 at 1:16 PM

Phil Laak plays poker as actress Jennifer Tilly keeps watch  (Source: University of Alberta)
Humans barely edge computer in Texas Hold 'Em tournament

At a tournament worth $50,000 held earlier this week  at the Association for the Advancement of Artificial Intelligence's annual conference in Vancouver, B.C., poker pros Phil "The Unabomber" Laak and Ali Eslami went up against Polaris, the reigning world champion computer-poker program.

The two poker pros were sat in different rooms, where each sat in front of a laptop to play Texas Hold ‘Em. The competition featured four matches, 500 hands each, between Polaris and the two poker playing professionals.

At the end of the first round, Polaris had a slightly higher chip count, though the total winnings difference was so small that the round was considered a draw (due to statistical variation). The second round – and the last for Monday – went in strong favor to Polaris.

The humans made a comeback the next day. At the end of the two day showdown, the two poker players came out $570 ahead – just enough to claim a small victory over the computer program.

"I really am happy it's over," said Eslami. "I'm surprised we won.... it's already so good it will be tough to beat in future."

The programmers behind Polaris used several different programs throughout its poker play to test which would be most effective. Phil Laak described one named Mr. Pink as a "careful, reasonable, disciplined, thoughtful player," while he called another one named Agent Orange as being "like a crazed, cocaine-driven maniac with an ax."

"The subtlety to the whole thing is, we won, not by a significant amount, and the bots are closing in," Laak told MSNBC. "That's the true summary."


Comments     Threshold


This article is over a month old, voting and posting comments is disabled
Those shades are going to help
By azrael201 on 7/26/2007 10:08:30 AM , Rating: 5
yea Laak, those sunglasses will prevent the laptop from calling your bluff




RE: Those shades are going to help
By ksherman on 7/26/2007 10:23:32 AM , Rating: 2
they forgot to mention that they use a camera built into the laptop to read the players :)


RE: Those shades are going to help
By Souka on 7/26/2007 10:42:04 AM , Rating: 2
and forgot to mention the value of chips in play...

$570 is nothing if they're playing $200,000+ buy ins, but quite a bit if they're doing $1000 buy ins.


By bunnyfubbles on 7/26/2007 1:39:15 PM , Rating: 2
true, I'd also like to know that, the only thing we have to go on is that the tournament was for $50,000...but that might not mean anything


RE: Those shades are going to help
By Alexstarfire on 7/26/2007 1:40:09 PM , Rating: 2
Considering how small a test this was, it could easily be said that either side won because of luck.


RE: Those shades are going to help
By grant2 on 7/26/2007 6:32:54 PM , Rating: 4
Considering how small a test this was, it could easily be said that either side won because of luck.

The same cards were dealt simultaneously in both matches to ensure luck was not an issue.

I.e., if the computer was dealt AA against Laak's KK in their hand #1, then Eslami was dealt AA vs. the computer's AA when he played HIS first hand.

2000 hands (500 per round x 4 rounds) isn't a huge amount, but it's certainly not a "small" test.


By damncrackmonkey on 7/26/2007 8:29:44 PM , Rating: 2
quote:
I.e., if the computer was dealt AA against Laak's KK in their hand #1, then Eslami was dealt AA vs. the computer's AA when he played HIS first hand.


I'm pretty sure that second part should have a KK in there. I'm guessing after "computer's."

Since the end victory was for the 2 poker players over the several artificial poker personalities, I think a statistical breakdown of each personality vs each player would be much more appropriate than just saying the humans won $570.

Anyway, since $570 is likely far less than the pot for any given hand, I don't think this can be taken as any evidence for pro poker players remaining better than computers.


RE: Those shades are going to help
By Samus on 7/26/2007 3:47:50 PM , Rating: 2
they were playing a $50,000 pot, $570 put them just over a 5% margin.


RE: Those shades are going to help
By fibbeh on 7/27/2007 11:33:04 AM , Rating: 2
They were playing $10/$20 limit hold'em. According to the official website, Phil and Ali won by a total of $395 after approximately 2000 duplicate hands or 4000 total hands. That means that the human players had a win rate of 0.5BB/100 (big bets per 100 hands) which is not a particularly high win rate at $10/$20 or lower, but wouldn't be bad at the actual levels that Phil Laak plays.
The "Pot" of $50,000 has no relation to the actual money they were playing with. Each player had $100,000 in "chips" to start with so that they would not run out of playing money in 500 hands. The research was only concerned with the net wins or losses by each player.


Reigning World Champion?
By Mattz0r on 7/26/2007 2:17:10 PM , Rating: 2
quote:
...Polaris, the reigning world champion computer-poker program...


Correct me if I'm wrong, but in order for Polaris to be a reigning world champion poker-playing program, wouldn't it have had to beat other poker-playing programs for the title?




RE: Reigning World Champion?
By Esteem on 7/26/2007 4:34:58 PM , Rating: 2
they did a machine only contest


RE: Reigning World Champion?
By Scrogneugneu on 7/26/2007 8:43:32 PM , Rating: 2
Now THAT is a good show.


This is a dumb test!
By ryedizzel on 7/26/2007 7:27:05 PM , Rating: 3
I play online poker at least 10 hours a day, and let me tell you something. Playing at a full table with 9 people is completely different than playing 1v1 poker. And I don't care what anyone says, a certain degree of Poker is about luck, especially with the river card. Even if I know what cards the other person is holding there is still no way of knowing what cards the dealer is going to flop next. In other words your pocket Aces can easily become crap if someone with a 2-7 flops 3 of a kind. I see it happen all the time...




RE: This is a dumb test!
By fibbeh on 7/27/2007 11:35:51 AM , Rating: 2
For this very reason, they dealt the exact same cards to Polaris vs. Laak and Ali vs. Polaris to eliminate as much of the short term luck as possible. This was also stated in a post above.


Amazing: is it really?
By FS on 7/26/2007 12:33:44 PM , Rating: 2
quote:
Man Emerges Victorious in Poker Match Against Machine


wasn't the machine designed by men(women) too.




Enterprise
By Ringold on 7/26/2007 1:05:10 PM , Rating: 2
If I were an enterprising young intern or the sorts with access to these bots.. I think I'd have to stick one on a laptop of my own, take it home, and sign up for a Poker site on the assumption most gambling gamers there wouldn't be as good as 'The Unabomber'.

I never said it'd be legal according to the sites rules probably. I just said I'd do it.




Big deal......
By marvdmartian on 7/26/2007 2:46:56 PM , Rating: 2
I mean, come on........

Hasn't just about everyone here gotten the girl naked in at least one game of online strip poker??? ;)




the REAL shocker is...
By UnFaZeD on 7/26/2007 10:22:55 PM , Rating: 2
Jennifer Tilliy 48....




"I'm suprised we won...."
By fibbeh on 7/27/2007 11:42:16 AM , Rating: 2
quote:
"I really am happy it's over," said Eslami. "I'm surprised we won.... it's already so good it will be tough to beat in future."


He may be surprised because of his matches vs. Polaris:

$10/$20 Limit Texas Hold'em - Ali Eslami
Session 1 : +$395
Session 2 : -$2495
Session 3 : -$635
Session 4 : +$460
TOTAL : -$2275




Next round will go different...
By oTAL (blog) on 7/27/2007 1:16:00 PM , Rating: 2
I believe next year things will go different.
I'll repost my comment from the article announcing the competition since I believe it's pertinent.

http://www.dailytech.com/article.aspx?newsid=7654

quote:

(...) Poker isn't like chess. There are two main things that make it very different to program:
1 - You don't know everything. The opponents cards are a mystery.
2 - Neither of you knows which cards will be placed on the table.

I distinguish both these factors since they must be handled differently.

A program can easily learn statistics and know the odds of any card being drawn and which hand is most probable to win.

On the other hand, it is a lot harder to effectively learn how a certain opponent reacts to a good or a bad hand or how he adapts his play to the amount of chips each player has and the size of the pot on the table. It is possible. It can even learn how to use bluffs and when to call or fold. But this is extremely hard to program and is compounded by the fact that you must "reward" the computer when it looses if the decision he made had the best probability of success (and "punish" him, even when he wins against the odds). Furthermore, you can't implement a decent reinforcement learning algorithm if you don't see your opponent's cards when any of you fold (this is a BIG deal since this will make it much harder for you to understand your opponent and evaluate your decisions). In training the computer should probably cheat and see the opponents cards when anyone folds.
Plus, even if they make the perfect program, but they make it deterministic (= same situation->same output), the opponents will find it easier to predict on certain situations. Obviously this isn't simple since the opponent does not know your cards, but it can be exploited in some situations. That means a small amount of randomness (or intolerance towards repetition) should be introduced on certain decisions.

This is a very difficult problem and this first time around, I would bet on the humans.
By next year, if the team is committed to the problem, I would bet on the computer. Why? Because the problem here is finding and fine tuning the best algorithms. The problem with chess is mostly lack of computer power for the needed calculations. You need algorithms that provide the best moves without taking years to compute. In poker you don't require that much computing power; only good algorithms, a very good implementation of reinforcement learning, and lots of time to play against the program (and to make it play against other computers).
The current situation is that a chess game can be lost because computers aren't fast enough, while a poker game will be lost because the algorithms weren't good enough (or simply bad luck since there's randomness involved). The first problem will not be easy to solve since you need massive improvements in computing power. The second can be solved by a single brilliant person.

Obviously, when you're at the tournament you will also require a bit of luck.




"There is a single light of science, and to brighten it anywhere is to brighten it everywhere." -- Isaac Asimov








botimage
Copy