Man Vs Machine Update...

[jdcox32 0]

http://www.cs.ualberta.ca/~games/poker/man-machine/Live/

[mase_gotsem 0]

computers winning so far thnx for the link

[jdcox32 0]

seems like the computer is playing super aggressive. it will be interesting to see the complete hand break down to see how the computer played a hand vs the way to human played the same hand. and the guy who is updating the blog is doing a pretty crappy job of it.

[rogerwilco 0]

In my opinion it is only a question of time before someone writes a program that plays poker better than most humans - if it doesn't exist already. The only significant problem I see is, whether or not the programmers themselves understand enough about the game to make a program that can beat a pro like Laak.

[AcesOnFire 0]

In my opinion it is only a question of time before someone writes a program that plays poker better than most humans - if it doesn't exist already. The only significant problem I see is, whether or not the programmers themselves understand enough about the game to make a program that can beat a pro like Laak.

What happens is that you get very good players collaberating with very good programmers.Yes there are already bots out there playing on multiple tables and crushing them.

[DrZoidberg 0]

Considering that just about anyone who has read a single article on low stakes NLHE has an edge on the majority of the fish out there, it's no surprise that there's probably hundreds of bots out there grinding away, if I could play real snug on like 100 low stakes tables 24 hours a day i'm sure I'd be rolling in virtual dough.

[Jeepster80125 0]

In my opinion it is only a question of time before someone writes a program that plays poker better than most humans - if it doesn't exist already. The only significant problem I see is, whether or not the programmers themselves understand enough about the game to make a program that can beat a pro like Ivey.

FYP

[Ecclesbury 0]

I'm no computer man, but surely they can NEVER build a computer program that could beat any humans in the long run?Not unless it had genuine Artificial Intelligence (or should I say genuine inteligence) but then we would have some Arnie "I'll be Back" crazy **** on our hands.Since any artificial intelligence has to follow a set of rules input by the programmer, surely the computer will follow the same betting procedures (even if they are randomised, they will be randomised with a certain percentage, i.e. raise aces 80% of the time and limp 20% of the time.)What I'm trying to say is that if a good enough human sat and played with the computer for long enough they could fingure out the computers betting patterns almost exactly and play accordingly.I suppose the question is whether there is a human good enough to do that, and whether you could write a program that would take a human more time/hands to figure out than they had in their lifetime?I think I've answered my own quetion.Huzzah.

[rogerwilco 0]

FYP

Well, following that logic, if you assume Ivey is the best there is, then only Ivey could make a program that can beat Ivey - beating Laak might be easier to achieve.

I'm no computer man, but surely they can NEVER build a computer program that could beat any humans in the long run?Not unless it had genuine Artificial Intelligence (or should I say genuine inteligence) but then we would have some Arnie "I'll be Back" crazy **** on our hands.

There are chess programs that not only beat most humans, but beat even the best humans in the game - a game that has been seriously played on a very high level for at least a couple centuries. I doubt that poker is really more complex than chess in that regard.ETA: And to make money with a poker-bot you don't even need a program that can beat professional players - beating $20 STTs, 3/6 LHE, 0.5/1 NLHE, etc. would easily be enough to make shitloads of money - and that can't be too difficult.

[Ecclesbury 0]

There are chess programs that not only beat most humans, but beat even the best humans in the game - a game that has been seriously played on a very high level for at least a couple centuries. I doubt that poker is really more complex than chess in that regard.

Good point, I was vaguely aware of that fact. I seem to remember as a kid when I was into chess a bit I heard a story about either Gary Kasparov (sp?) being the first human to beat a certain computer program, or a computer program being the first to beat Kasparov, I forget which.

[hblask 1]

Since any artificial intelligence has to follow a set of rules input by the programmer, surely the computer will follow the same betting procedures (even if they are randomised, they will be randomised with a certain percentage, i.e. raise aces 80% of the time and limp 20% of the time.)

One of the poker books I read (can't recall which one right now) discussed the mathematics of randomizing your bets. The net result was basically that if you can correctly randomize your betting patterns, then no matter what action your opponent takes, it is -EV to them. I'll see if I can figure out which book it was.

[KONGOS 0]

Well, following that logic, if you assume Ivey is the best there is, then only Ivey could make a program that can beat Ivey - beating Laak might be easier to achieve. ;)There are chess programs that not only beat most humans, but beat even the best humans in the game - a game that has been seriously played on a very high level for at least a couple centuries. I doubt that poker is really more complex than chess in that regard.ETA: And to make money with a poker-bot you don't even need a program that can beat professional players - beating $20 STTs, 3/6 LHE, 0.5/1 NLHE, etc. would easily be enough to make shitloads of money - and that can't be too difficult.

http://www.msnbc.msn.com/id/19875057/This is what makes poker interesting...“You don’t have perfect information about what state the game is in, and particularly what cards your opponent has in his hand,” said Dana S. Nau, a professor of computer science at the University of Maryland in College Park. “That means when an opponent does something, you can’t be sure why.”As a result, it is much harder for computer programmers to teach computers to play poker than other games. In chess, checkers and backgammon, every contest starts the same way, then evolves through an enormous, but finite, number of possible states according to a consistent set of rules. With enough computing power, a computer could simply build a tree with a branch representing every possible future move in the game, then choose the one that leads most directly to victory."

[pele_br 0]

Well, following that logic, if you assume Ivey is the best there is, then only Ivey could make a program that can beat Ivey - beating Laak might be easier to achieve. ;)There are chess programs that not only beat most humans, but beat even the best humans in the game - a game that has been seriously played on a very high level for at least a couple centuries. I doubt that poker is really more complex than chess in that regard.I would disagree with your statement that poker is not more complex than chess. I think that the dynamics of the two games are very different, especially since both players "know" what the other player is holding at all times (both players can see the other's pieces, and positioning). The same is not true for poker, where, bluff is a big part of the game. A perfect example is Annette15's video where she is playing blind and still manages to win an MTT, I doubt a computer could achieve similar results.ETA: And to make money with a poker-bot you don't even need a program that can beat professional players - beating $20 STTs, 3/6 LHE, 0.5/1 NLHE, etc. would easily be enough to make shitloads of money - and that can't be too difficult.

The above statement is probably correct IMO though...

[telescop 0]

Good point, I was vaguely aware of that fact. I seem to remember as a kid when I was into chess a bit I heard a story about either Gary Kasparov (sp?) being the first human to beat a certain computer program, or a computer program being the first to beat Kasparov, I forget which.

Chess is VERY different in the fact that the computer has complete information about the game and in poker the computer doesn't know your cards.EDIT:^^^^^^^^^^^WHAT HE SAID UP THERE

[DrZoidberg 0]

anyone else find it slightly odd that there was a tie? wtf?

[TheMathProf 0]

There's a tie anytime one team wins by less than 25 small bets, counting it as statistical noise. The computer "won" the tied first set by 7 small bets.Edit: I should look up that information for real before I answer.

[jethrodull 0]

There's a tie anytime one team wins by less than 25 small bets.

fyp

[bigbrennan 0]

You cannot compare a program that plays chess to a program that plays poker.Chess a program simply analyzes each possible move and 5 (or so) moves after, and gives that option a score. Then will follow the highest scoring option.You can't do that with poker. As poker (in programing terms) is based heavily on luck it is very hard to code a system which will understand the situation.Do you assume the range of your opponents? Or do you teach it to gather intelligence every time there is a show down?Phil Ivey starting range is different to that of Phil Laak. Do you feed the software with pre collated data on each player? But doing this you are not letting the computer 'play' but instead take educated guesses at what it should do.

[rogerwilco 0]

I would disagree with your statement that poker is not more complex than chess. I think that the dynamics of the two games are very different, especially since both players "know" what the other player is holding at all times (both players can see the other's pieces, and positioning). The same is not true for poker, where, bluff is a big part of the game. A perfect example is Annette15's video where she is playing blind and still manages to win an MTT, I doubt a computer could achieve similar results.

I understand that argument, and I'm not saying it's wrong, but I don't completely agree either. Analysing betting patterns and finding good places to bluff is not impossible to write into a computer program. Making marginal bets and calls after pinning down your opponent's hand ranges while playing mathematically perfect is not impossible to do for a computer program. Something that probably is very difficult to incorporate are psychological factors, but most other aspects of poker can be done, imo. A program doesn't need to get everything right to have an edge against most players.Also, I'm not good at chess, but it is a game where creativity and intelligence play a role, it is not just about making "the right move" and thinking a couple moves ahead.I guess my point is: You don't need a real AI for a poker program to be better at that game than a vast majority of humans, that it is just a matter of programming (and not even a very difficult one).

Phil Ivey starting range is different to that of Phil Laak. Do you feed the software with pre collated data on each player? But doing this you are not letting the computer 'play' but instead take educated guesses at what it should do.

Humans playing poker do that too!

[KONGOS 0]

I'd like to see the results of a NL game and how the computer would play against someone like Townsend, Ivey etc.

[Spademan 94]

The biggest advantage a computer will have against humans is a mixture of no fatigue and no tilt.

[bigbrennan 0]

The biggest advantage a computer will have against humans is a mixture of no fatigue and no tilt.

unless you bump into its desk...

[gfdsa146 0]

is there any videos of this??? I would have loved to see Phil Laak when he value-bet 2nd bottom pair on the river to get called by Polaris' bottom pair. It said he went crazy!

[juggalo41729 0]

they're playing now live updates on that link in the op.

[takemychips 0]

is there any videos of this??? I would have loved to see Phil Laak when he value-bet 2nd bottom pair on the river to get called by Polaris' bottom pair. It said he went crazy!

I just spent 30 minutes in the hotel watching Phil and he's keeping up a decent commentary of his actions. He's even keeping the crowd involved... reviewing the hands with the audience and responding to their comments. Also saw Jennifer Tilley there briefly...