can you imagine phil hellmuth playing against libratus? "it called with ace-ten honey!" "this computer is the worst poker player I've ever seen! really it probably won't last another hour!"
I wonder how the study could go if for the first day you tell the players they are playing against an anonymous pro who is and in another room? I feel there can be a placebo effect with players being conscious of the AI. This was a very interesting report, thanks for posting!
Poker is still a game of luck. All the AI is doing is trying to approximate the Nash equilibrium strategy as closely as possible. It doesn't win by being "good" it wins by making less mistakes than humans
Nullpointer but if it wins it is the better player. The better player makes fewer mistakes. It (the AI) doesn't get tired, frazzled, pissed, euphoric or impulsive.
They are not the best HU players in the world tho.. Which kinda sucks, would be great to see the comp vs the best of the best. They're very good pros.. but best HU they are not.
tfw this is actually fake news too... The test is flawed, they are playing mirrored hands and are equity chopping all in scenarios which fundamentally changes the game and the way mirrored hands should work. Dong Kim explains this in his interview with Doug Polk
I am really interested in seeing how Libertus adapt his play when facing itself. I can reason that it would reach several stationary states all of which are only metastable. The issue is how can it closely approach the optimal strategy while the other Liberatus is doing the same. Also it is amazing how easily Liberatus drifts far away from GTO to exploit the opponent! Great job!
I forsee a future where a bearded Poker-playing AI with chicken leggs, is posting Instagram pictures filled with hired models, guns and luxurious mansions. And people will like the AI for that.
The most interesting bit is that, if the network was trained against a computer algorithm it would learn the loopholes embedded within its syntax. When trained against a human it is learning our flaws and mannerisms. I say we probably shouldn't strive for a general A.I unless we wish to be fully displaced.
Yep, AI can make very weird moves that you rarely see high level players use, it's just because they can go through every scenario that pro level players don't consider due to an undiscovered way of thinking. Chess, poker, whatever. Any player that played with an AI can notice a difference.
It's tue that old style AI's (hard-coded logic with future-move analysis) often have really weird moves. But modern Neural-network AI's do learn like humans (from other humans, through experience, just at a really high rate) and have much more human-like moves. In fact, AlphaZero, the version of the Alpha AI that leans without being fed human game replays, has rediscovered many human strategies by itself. AI's in much more complex real-time games (like DOTA and StarCraft) do also play a bit like humans (although they feel oddly smart and skilled), and show human-like behaviour like faking, baiting, smart exploiting of game mechanics,... But when they feel inhuman, it is mostly because they have much more experience, they have seen and played millions of games.
1. Those players are not "absolute best". Far from it 2. The player hadn't had the usual incentive to play to win. 3. The players chose to play as a marathon (being tired), to get the same money in less days 4. It's a statistical tie (insider joke)
They are very good players but not the "best" in the world. Also, the computer can not predict what cards are coming out next. Nor would it know what you mucked. Also, the computer has a bit of an advantage because it can instantly calculate odds that would be too complicated for a human, especially since humans get tired. I would like to see it play against a mix of regular players and much higher level players. Also, the computer is playing unfairly. Human plays aren't bringing their charts, computers, calculators, or consulting outside sources. Level the playing field and let the humans use any resources they want which would cut down of fatigues and human error.
@@jburch5752 I think they were using some tools to analyze what was working and not working but yes you are right computers can do calculations in second. I'm reading the poker blueprint and the formula to find ev is too long to do each hand. That's why we have to just estimate and review later unlike computers
This is a monumentous task involving having to vary your strategy over time and even changing the rate at which you change strategies over time, whilst also having each style be concrete enough in order to not leave too much up to chance. All the while, the AI's model has a discrete memory. Since it's playing against multiple people and learning from them whilst still keeping individual strategies for individual players, they are playing against a machine with the solid foundations of many which can focus that knowledge in order to wreack havoc. Like a prized warrior faced by the collective efforts of a small army, terrifying and also very instructive and awe inspiring!
5:20: How did he arrive at that figure? Don't get me wrong, the number of possible situations that can occur in a full ring game are incalculable, but the amount of possible permutations in a HU game is roughly 5.75 quadrillion.
I would like to see if there would be any variance if the cards were real and the info of the hand was inputted into the Libratus instead of playing the game in the computer that has Libratus on it.
I'm wondering how this AI would do in 6 handed sit&go's for example. I'll bet that it won't be long before most poker sites will use a version of this AI to consolidate and probably even increase their margins. And I'm wondering how long it will before an AI will win the WSOP using 'speech play'?!! Like a boss!!
6max sitngoes can be beaten by bots pretty easily. However, a bot beating a 6max cash game table over 100k hands would be extremely hard for a bot if all other 5 players were extremly talented (such as the ones in this video). It would be even harder to beat if it was a 9max cash table.
The theoretical problem with a 6 handed cash table is there are exponentially more situations that the bot could come across. Logically it would mean that they need a lot more computing power to pull it off than they do even now!
As for the best strategy for improving odds, considering the ai has better statistical data, strategy, and is overall steps ahead of the player, I think the best odds you have are going all in on premium/high valued probability hands, as the players started to do. The ai will still beat you at valueing it's hand, but by making it all in every time, it removes the river, flop, and turn betting strategies, and puts more into preflop odds, as well as limit the ammount of hands the game is played, limiting how much the statistics effects the outcome. Although, the ai would win over 50%, I presume this is the closest you could get it to even win/loss. Simply do your best to remove all strategy from the game, and put your chips in on luck.
Have a computer on the side playing online poker and each hand you play just enter the exact cards you have and the flops then let the Ai tell you what to do. Mark witch position is checking or raising and let the computer give you the next move. (Example, fold because that that that or 3 bet because i see weakness, you will win millions) I’m programming my own poker AI assistant for online games
So many people commenting that the ai is cheating because the "computer" knows the cards ... are people really that dense they think 5 pro poker players and multiple leading scientists didn't think about that. I'm pretty sure the ai didn't know the cards or his winrate would have been 95 %
last year their bot got crushed in almost the same way it did the humans this year but the company declared it a statistical tie in an official report. so gratz on a statistical tie.
Actually, heads up is harder than a full table. You have the luxury of waiting for cards on a full table. However, heads up, you are blinded in every hand. You are forced to outplay your opponent without good cards.
Faster game does not mean a harder game. More players means more variables, both known and unknown. There are other videos about this AI that explains that it doesn't do well against more than one opponent ....yet.
More players the more variables. For the computer playing heads-up is far easier. Example, pocket aces heads up is far stronger than it would be in a 10 handed game.
I understand that they're probably more interested in the practical applications of this AI, but I would LOVE to have Libratus play itself for thousands of hours until it converges (or comes close) to the Nash Equilbirum for HUNL. It would really improve the level of human players, in much the same way that chess engines have vastly improved the play of chess professionals.
So what would happen if you pitted this algorithm against itself? Would it basically never converge and just have the strategies oscillating around equilibrium?
i am just saying that the raw computing power that is necessary to solve HUNL is just not there yet. As the video says, there are like 10^160 scenarios and therefore a huge number of game trees each with different possible bet sizings etc.
Finding the Nash equilibrium for Texas Holdem is like finding the perfect strategy (God's hand) for chess or Go; it's impossible with a computer smaller than the observable universe. But I imagine it should be possible to go far beyond the best human player, albeit the result is less apparent since victory is still dependent on chance.
Not really, people can't afford to make those kinds of bots...atleast poker players can't and wealthy businessman wouldn't bother as they play for fun..also NL texas holdem is only 1 form of poker, there are far more complex game like PLO(Pot limit omaha) which is theoritically impossible to solve.
You live in a naive reality if you believe people with the financial ability to employ this technology in order to cheat ordinary folks out of money " wouldn't bother" . Go watch some video on how technology is used to provide fractional edge to traders in the stock market. The rich will pursue any, ANY, method to extract money from people. I can't emphasize the word " any " quite enough.
torgo4ever Think about it for a second, how many multi millionaires play poker ? Prob less than couple hundred, then they have to have the motive to cheat, which maybe let's give a optimistic guess around a dozen, then they gotta find someone to make such kind of technology and lastly even somehow couple people manage to do it they gotta worry about not getting banned or variations in formats being bought in which drowns millions of dollars which doesn't sound like a great businesses plan, see the flaw in your logic ? If they really wanna cheat they will find more reliable and cheaper scams.
torgo4ever And lastly this would only be at the highest stakes which only few dozen even play on weekends....99.95 poker population doesn't get effected by this...
10^160 poker combinations. "More combinations than there are atoms in the universe." This is true, but it's also a gross understatement. If you created a copy of the universe for every atom in our universe, *then* counted the number of atoms in all of them combined you'd be far closer.
Libratus is likely the best bot at this point, but you can check out the Anual Computer Poker Competition, where many good bots play against each other. Even an earlier bot from the makers of Libratus and other universities as well.
The comment at the end that AI is always helping mankind is what so many scientists/futurists are worried about. Programming AI with that kind of amorphous, subjective criteria is beyond difficult. Awesome story here, regardless.
I'm wondering if it took 20 days of poker play for Libratus to solve the game? Can it fair as well in a normal poker time frame like 3-8 hours of gameplay?
Does it learn from data specific to each player or that specific aggregate group and their various learn exploits? How would this scale if it was just a station game at a casino with no hand large hand history for whoever sat at the machine? Is it optimising for poker in general? or against the specific player and their data?
The way to beat Libratus is by not giving it enough hands or information to figure you out. You have to beat it quick and be almost random. Machines can't make sense out of utter randomness.
Yeah it is learning the players and what the players do, not what all players do. The same way completely new players can defeat seasoned players simply because they don't understand what they have or what the other person is pretending to have. Most professional players play the math/odds game. They bet certain amounts based on their cards to get payouts proportionately. The supercomputer outmaths them.
It would be intereesting to see a total rookie being told what to do by the program playing against a pro. I wonder if the pros could read the rookies.
Questions:- 1) Did the blinds increase over time or were blinds the same through out the competition? Could the computer account for the Blinds increasing? 2) Could the AI in its present form compete at a 10 person table tournament? 3) What other games left does google AI want to compete in?
1. The blinds stayed the same throughout. Also each hand started with both players having a 20k stack. So each hand both players were equally deep. 2. This is just a guess, but I wouldn't be surprised if it would do reasonably well. The team behind Libratus has said that in their estimates Libratus could crush 6-max in 2 years. I think this is a careful estimate, and believe it can be done sooner if they put their effort on doing it. 3. Your guess is as good as mine here.
Wayne Johnson the players are paired up and they get reversed hole cards to reduce variance. For example if Jason gets AA vs KK, Daniel will get KK vs AA.
I’m telling you these boys already had computers doing this online, I’m a great player, but when he said over bets and underbets, I get flashbacks to when these suspicious ids from sites would destroy me, I would always remember them cause they bet so random and just crazy , except it would work on me. There where only a few, but I’m sure I’ve played these bots that sites already have made.
Not even watching the entire video because the title already sums it up. It should be "Can AI beat the best poker players in the world?", then i'd be watching the entire vid.
lol, he's giving a lecture there where the tables are empty, literally no players are playing around 7 am, I'm probably one of the weirdos playing lmao
That's pretty cool and all, but it doesn't take into account one of the most important aspects of poker - the fear of losing. The computer is not afraid to lose and so that has no effect on the decisions it makes, while for the human players it is a pretty significant factor, even at higher stakes. So in that sense, it has an unfair advantage over its opponents.
It does not work like that at pro lvl even though winning or losing still affects everyone at some lvl. What i want to say is that pros are less affected by this because they went through countless swings of variance along they pro life. They just simply play the solution they find and do not bother with the result. If they know they are better they know that eventually the EV line will converge. In this situation they felt they are outplayed and not because of the result.
I guess I didn't express myself correctly when I said that fear has an even more significant impact at higher stakes. I agree with you that pro players are better at dealing with the pressure than regular players. That's one of the main reasons why they are more successful. However, even though they are less affected by the tension compared to normal people, they are still affected to some degree, especially when there are millions on the line. After all, they are still human and even the calmest, coolest person, sometimes can't help but feel a little doubt or insecurity under extreme pressure. I think that this is the main reason why there were outplayed - a human can never match a machine in this regard and so in a sense they were playing at a disadvantage from the get-go.
9:16 If I may add a slightly more grim approach. (First of all, I want to say that I think tool development is one of the best things to help human kind overall) But, the argument that he uses about having AI helping negotiations, by that point in time (and maybe, unfortunetly, it willl always happen), the "boss" will always have a better tool than the employee, specially for big corporations. Maybe progress will hault and become accesible, but we often have to think about the implications of our technological improvements.
i somehow believe overbetting turn card makes river plays much easier ! Ofc you need to be in possition/ALWAYS! If you're out of possition you should ALWAYS play trap game no matter if u AA or 72 holding.
These are not the best players in the world. They don't have much information about their success on the internet. They're playing 1500 hands a day which is an enormous amount, with 10 second shot clocks. Under those conditions, an OK professional player isn't going to be playing anywhere near the best. These players admitted to playing terrible strategies just hoping to find a weakness in the bot, rather than playing their own best strategies. The robot team is overstating their accomplishment here. I'd like to see this bot challenge the actual known best players in the world.
I still think a lineup of Dong Kim, Doug Polk, Daniel Cates, Ben Sulsky would have beat them. These guys were top 20 players but not the best other than Kim
I believe that Dong was considered the number two heads up no limit player in the world and he still had a slight loss against Libratus. Doug Polk is the only one who would've had a chance to beat Libratus.
I do believe there's a precise way of kicking the AI butt.... 10000% randomness ..... everything all over the map, so it can't figure out your strategies the same way you can't figure out its .... much closer result i bet.
can you imagine phil hellmuth playing against libratus? "it called with ace-ten honey!" "this computer is the worst poker player I've ever seen! really it probably won't last another hour!"
Mr Tomato what?
Arya Khonsari thats a Hellmuth quote
fuckin durrrrr
The AI probably can’t even spell Poker
Northern European hardware
I wonder how the study could go if for the first day you tell the players they are playing against an anonymous pro who is and in another room? I feel there can be a placebo effect with players being conscious of the AI.
This was a very interesting report, thanks for posting!
BigTeste my thoughts exactly. They went in trying to beat an bot, instead of trying to win the most money.
A robot plays very differently to a human. You’d be able to tell it was a robot within an hour or so even if nobody told you.
Who would like to see Phil hellmuth play against this AI? That be funny as hell I imagine
He called a raise with QT. Probably doesn't even know how to CODE poker.
Doppe1ganger "he 3 bet with Q 10 honey"
instantly laughed my ass off!
LMFAO "fucking moron AI, I had pocket tens!!"
haa. he would die from rage
this proves poker is a game of skill
peppi1974 u mean poker is game of feel at got no feel
bullshit its prove only one thing - that time comp was lucky
Poker is still a game of luck. All the AI is doing is trying to approximate the Nash equilibrium strategy as closely as possible. It doesn't win by being "good" it wins by making less mistakes than humans
Nullpointer but if it wins it is the better player. The better player makes fewer mistakes. It (the AI) doesn't get tired, frazzled, pissed, euphoric or impulsive.
@@bulbarobat Lucky? For hours?
Talking about the latest technology in AI while sitting in front of a CRT monitor...
+momentinpassing yeah, but CRT actually means Cathode Ray Tube, which is also cool. And by the way, they don't work with x-rays.
Maybe he's a smash melee fan
AI strategy is both amazing and kind of terrifying.
This was a great report. Always happy to see journalism at its finest in the age of fake news and improper journalism.
Engadget is a web magazine. Magazines are under the umbrella of journalism.
They are not the best HU players in the world tho.. Which kinda sucks, would be great to see the comp vs the best of the best. They're very good pros.. but best HU they are not.
Mr J Bean I completely agree with you
Indeed. But pewdiepie still has 150 times more subs.
tfw this is actually fake news too... The test is flawed, they are playing mirrored hands and are equity chopping all in scenarios which fundamentally changes the game and the way mirrored hands should work. Dong Kim explains this in his interview with Doug Polk
Kassouf could make Libratus tilt by hand 4.
I am really interested in seeing how Libertus adapt his play when facing itself. I can reason that it would reach several stationary states all of which are only metastable. The issue is how can it closely approach the optimal strategy while the other Liberatus is doing the same. Also it is amazing how easily Liberatus drifts far away from GTO to exploit the opponent! Great job!
I forsee a future where a bearded Poker-playing AI with chicken leggs, is posting Instagram pictures filled with hired models, guns and luxurious mansions. And people will like the AI for that.
Hahahaha omg that so on point.
The most interesting bit is that, if the network was trained against a computer algorithm it would learn the loopholes embedded within its syntax. When trained against a human it is learning our flaws and mannerisms. I say we probably shouldn't strive for a general A.I unless we wish to be fully displaced.
Luke Bitton tell prometheus that
Luke Bitton EXACTLY. Do they not see the simple and almost certain future in creating this??
Yep, AI can make very weird moves that you rarely see high level players use, it's just because they can go through every scenario that pro level players don't consider due to an undiscovered way of thinking.
Chess, poker, whatever. Any player that played with an AI can notice a difference.
well said, some of the chess moves those machines play are obvious machine moves, !
It's tue that old style AI's (hard-coded logic with future-move analysis) often have really weird moves. But modern Neural-network AI's do learn like humans (from other humans, through experience, just at a really high rate) and have much more human-like moves.
In fact, AlphaZero, the version of the Alpha AI that leans without being fed human game replays, has rediscovered many human strategies by itself.
AI's in much more complex real-time games (like DOTA and StarCraft) do also play a bit like humans (although they feel oddly smart and skilled), and show human-like behaviour like faking, baiting, smart exploiting of game mechanics,...
But when they feel inhuman, it is mostly because they have much more experience, they have seen and played millions of games.
Hello Engadget, I would like to look at the played hands. Is it possible to see a record?
I like how he keeps calling them humans
hahaha
1. Those players are not "absolute best". Far from it
2. The player hadn't had the usual incentive to play to win.
3. The players chose to play as a marathon (being tired), to get the same money in less days
4. It's a statistical tie (insider joke)
I was thinking some of these exact things
Denial
@@isaacvongurtberg7341 Valid tagging while not even trying to refute, PUSSY
They are very good players but not the "best" in the world. Also, the computer can not predict what cards are coming out next. Nor would it know what you mucked. Also, the computer has a bit of an advantage because it can instantly calculate odds that would be too complicated for a human, especially since humans get tired. I would like to see it play against a mix of regular players and much higher level players. Also, the computer is playing unfairly. Human plays aren't bringing their charts, computers, calculators, or consulting outside sources. Level the playing field and let the humans use any resources they want which would cut down of fatigues and human error.
@@jburch5752 I think they were using some tools to analyze what was working and not working but yes you are right computers can do calculations in second. I'm reading the poker blueprint and the formula to find ev is too long to do each hand. That's why we have to just estimate and review later unlike computers
"Genomic research, improving society..." LIBRATUS..Libr-rat-us... Liberate Us?... Yea, this is some WestWorld shit.
Libratus means "balanced" in Latin.
Lab Rat Us.
tbh i thought it had something to do with liberation as well
This is a monumentous task involving having to vary your strategy over time and even changing the rate at which you change strategies over time, whilst also having each style be concrete enough in order to not leave too much up to chance. All the while, the AI's model has a discrete memory. Since it's playing against multiple people and learning from them whilst still keeping individual strategies for individual players, they are playing against a machine with the solid foundations of many which can focus that knowledge in order to wreack havoc. Like a prized warrior faced by the collective efforts of a small army, terrifying and also very instructive and awe inspiring!
Libratus should "check its privilege"
hahaha
untouchable360x lololol
This sh*t is going to destroy online poker
Ruben Ernst once a supercomputer becomes available to the general population, sure
mehdinadif you don't need a supercomputer to run AI software
Manuel LK check the video, liberatus specifically needs a supercomputer to perform its calculations.
It doesn't work in anything apart from heads up NL. There is no nash equaliblium in 6max.
Ruben Ernst going to destroy humanity
I'm sure this greatly improved those HU players game.
I'm sure this AI is somehow sexist, but I haven't figured out how exactly..
The AI is actually a feminist
Guys please try to remember that full ring NLHE is still very far from being solved.
AI beat vs 5 pros in a 6 max game
"just in libratus fashion" lmfao how many times did the robot suckout loooool
5:20: How did he arrive at that figure? Don't get me wrong, the number of possible situations that can occur in a full ring game are incalculable, but the amount of possible permutations in a HU game is roughly 5.75 quadrillion.
One of the biggest advantages to AI. There is no fear of losses or mistakes.
Put Ivey in there. I'm serious
I discovered a good compilation of videos that should help on card crusher fixer
I agree; Ivey is not of this world!
Wait do any of you think there is a fighting chance if ivey play the AI?
The top online players would shit all over Ivey.
@@futurez12 No they wouldn't
I would like to see if there would be any variance if the cards were real and the info of the hand was inputted into the Libratus instead of playing the game in the computer that has Libratus on it.
The Ai are or is reading the cards on the screen through your eyes is how it improves it moves each game.
I'm wondering how this AI would do in 6 handed sit&go's for example.
I'll bet that it won't be long before most poker sites will use a version of this AI to consolidate and probably even increase their margins.
And I'm wondering how long it will before an AI will win the WSOP using 'speech play'?!!
Like a boss!!
"How many miles from here to silicon valley?"
InfiniteCyclus 6max sngs can easily be besten by bots Today. Its basically just math, Cashgame HU is a complete differentierad story
6max sitngoes can be beaten by bots pretty easily. However, a bot beating a 6max cash game table over 100k hands would be extremely hard for a bot if all other 5 players were extremly talented (such as the ones in this video). It would be even harder to beat if it was a 9max cash table.
its not programmed for that,.,. im assuming its way more deepstacked so more skill intensive for 6handed lowskill game it would be different bot
The theoretical problem with a 6 handed cash table is there are exponentially more situations that the bot could come across. Logically it would mean that they need a lot more computing power to pull it off than they do even now!
So the professor himself coded and thought out the whole project himself? That would be a first.
As for the best strategy for improving odds, considering the ai has better statistical data, strategy, and is overall steps ahead of the player, I think the best odds you have are going all in on premium/high valued probability hands, as the players started to do. The ai will still beat you at valueing it's hand, but by making it all in every time, it removes the river, flop, and turn betting strategies, and puts more into preflop odds, as well as limit the ammount of hands the game is played, limiting how much the statistics effects the outcome. Although, the ai would win over 50%, I presume this is the closest you could get it to even win/loss. Simply do your best to remove all strategy from the game, and put your chips in on luck.
Have a computer on the side playing online poker and each hand you play just enter the exact cards you have and the flops then let the Ai tell you what to do. Mark witch position is checking or raising and let the computer give you the next move. (Example, fold because that that that or 3 bet because i see weakness, you will win millions) I’m programming my own poker AI assistant for online games
So many people commenting that the ai is cheating because the "computer" knows the cards ... are people really that dense they think 5 pro poker players and multiple leading scientists didn't think about that. I'm pretty sure the ai didn't know the cards or his winrate would have been 95 %
Poker sites could offer this as an opponent as a sort of "Beat the AI" challenge.
Congrats to humans on statistical tie
Dude. Not a tie. The humans got crushed.
last year their bot got crushed in almost the same way it did the humans this year but the company declared it a statistical tie in an official report. so gratz on a statistical tie.
Definitely not a statistical tie. Humans lost big time to the bot.
well, did you see the results of the last years challenge? when that was a statistical tie this is also one.
Sorry, dude. Not how it works. Getting crushed is getting crushed.
You should challenge Libratus against himself, would be interesting
It played itself 100 trillion times. That's how it was created, the losing strategy trees were cut quickly.
5:33 it explains it.
Reinforced learning brah
This will kill online Poker in Future... I am pretty sure...
How..? Maybe online poker but not live
@@gonssavm4048 re read his comment again.
@@PartyRaveAnimals lmao my bad
@@gonssavm4048 its okay lol
It's already killed
It's a crime that Noam Brown was not mentioned anywhere in the video or the description.
Heads up poker is pretty non-standard and heavily based on luck. I'd be more impressed seeing it win at a full table.
Actually, heads up is harder than a full table. You have the luxury of waiting for cards on a full table. However, heads up, you are blinded in every hand. You are forced to outplay your opponent without good cards.
Faster game does not mean a harder game. More players means more variables, both known and unknown. There are other videos about this AI that explains that it doesn't do well against more than one opponent ....yet.
More players the more variables. For the computer playing heads-up is far easier. Example, pocket aces heads up is far stronger than it would be in a 10 handed game.
I understand that they're probably more interested in the practical applications of this AI, but I would LOVE to have Libratus play itself for thousands of hours until it converges (or comes close) to the Nash Equilbirum for HUNL. It would really improve the level of human players, in much the same way that chess engines have vastly improved the play of chess professionals.
solving HUNL or coming close to solving it is def not possible with the current state of computer power
So what would happen if you pitted this algorithm against itself? Would it basically never converge and just have the strategies oscillating around equilibrium?
i am just saying that the raw computing power that is necessary to solve HUNL is just not there yet. As the video says, there are like 10^160 scenarios and therefore a huge number of game trees each with different possible bet sizings etc.
Finding the Nash equilibrium for Texas Holdem is like finding the perfect strategy (God's hand) for chess or Go; it's impossible with a computer smaller than the observable universe. But I imagine it should be possible to go far beyond the best human player, albeit the result is less apparent since victory is still dependent on chance.
I'm pretty sure Libratus playing trillions of hands against itself is how it devised its initial strategy vs. the humans.
this is the end of online poker. I am sure some people will continue to play but they'll be nothing more than donation jars
Not really, people can't afford to make those kinds of bots...atleast poker players can't and wealthy businessman wouldn't bother as they play for fun..also NL texas holdem is only 1 form of poker, there are far more complex game like PLO(Pot limit omaha) which is theoritically impossible to solve.
You live in a naive reality if you believe people with the financial ability to employ this technology in order to cheat ordinary folks out of money " wouldn't bother" . Go watch some video on how technology is used to provide fractional edge to traders in the stock market. The rich will pursue any, ANY, method to extract money from people. I can't emphasize the word " any " quite enough.
torgo4ever Think about it for a second, how many multi millionaires play poker ? Prob less than couple hundred, then they have to have the motive to cheat, which maybe let's give a optimistic guess around a dozen, then they gotta find someone to make such kind of technology and lastly even somehow couple people manage to do it they gotta worry about not getting banned or variations in formats being bought in which drowns millions of dollars which doesn't sound like a great businesses plan, see the flaw in your logic ? If they really wanna cheat they will find more reliable and cheaper scams.
torgo4ever Also it would take decades for them to make such a sophisticated bot.
torgo4ever And lastly this would only be at the highest stakes which only few dozen even play on weekends....99.95 poker population doesn't get effected by this...
where do i download libratus?
online poker is screwed. VR poker is the next big thing.
ive thought this for a long time, you have to be able to webcam to play with captcha popups like some sites have now
That AI is like pokerstars micro players.
In this poker game, is there a cutoff point where you know the average of hands is better or worse than what you currently have?
well to be fair they dont call the best HU player in the world dan polk
*Doug Polk
Go Donger Kim with the smallest loss!
10^160 poker combinations.
"More combinations than there are atoms in the universe."
This is true, but it's also a gross understatement. If you created a copy of the universe for every atom in our universe, *then* counted the number of atoms in all of them combined you'd be far closer.
Playing the AI is probably like playing against Shang Tsung in Mortal Kombat morphing into whatever style suits it better at the time.
What if you make the AI vs the AI ?
It played against itself trillions of times, constantly adjusting its algorithms. That's how it got so good.
Daniel F boom mind blown
Would be fun to watch to be fair
Libratus is likely the best bot at this point, but you can check out the Anual Computer Poker Competition, where many good bots play against each other. Even an earlier bot from the makers of Libratus and other universities as well.
Computer explode
Behind Libratus there is a real person. You can his picture on the top-left of the screen at 10:44
As a poker player, I wish I could play this AI! 😁
Try a 6-max THNL AI player
The comment at the end that AI is always helping mankind is what so many scientists/futurists are worried about. Programming AI with that kind of amorphous, subjective criteria is beyond difficult. Awesome story here, regardless.
I think if players could see libratus' hands everytime it would be easier for them to adapt
Does this system use players hole card information to collect this data or just what it learns from show downs?
Be interesting to see these sessions cards up
Poker is not just about the maths. It's about the human interaction as well, and what makes it interesting.
A win win for all of us great work.
I'm wondering if it took 20 days of poker play for Libratus to solve the game? Can it fair as well in a normal poker time frame like 3-8 hours of gameplay?
About 14 seconds into the video made my siri go off? Anyone else experience this?
My "hey siri" is turned on and my iphone is plugged in.
Okay, now play 100,000 big blinds deep. =)
Makes it even harder for humans
Does it learn from data specific to each player or that specific aggregate group and their various learn exploits? How would this scale if it was just a station game at a casino with no hand large hand history for whoever sat at the machine?
Is it optimising for poker in general? or against the specific player and their data?
Lol Phil helmuth stands up at the computer, rips out his headphones. Yells at the computer "you are a idiot friend" lol
how can i download this?
The way to beat Libratus is by not giving it enough hands or information to figure you out. You have to beat it quick and be almost random. Machines can't make sense out of utter randomness.
Sb Espn this is no true lmao. Don't try this people
Yeah it is learning the players and what the players do, not what all players do. The same way completely new players can defeat seasoned players simply because they don't understand what they have or what the other person is pretending to have. Most professional players play the math/odds game. They bet certain amounts based on their cards to get payouts proportionately. The supercomputer outmaths them.
Libratus plays an approximated Nash equilibrium, it doesn't have to read people.
Who deals the cards in beginning
I want to see an AI developed to beat the best TCG players.
AI flips over set Heavy Storm
Really interesting. It would be interesting to see if Doug Polk, Isaac Haxton, Jonas Mols and Alex Millar could beat it
It would be intereesting to see a total rookie being told what to do by the program playing against a pro.
I wonder if the pros could read the rookies.
Questions:-
1) Did the blinds increase over time or were blinds the same through out the competition? Could the computer account for the Blinds increasing?
2) Could the AI in its present form compete at a 10 person table tournament?
3) What other games left does google AI want to compete in?
misomiso It won't be a tournament format It'll be a cash format with less variance. They may play at different stack sizes though
1. The blinds stayed the same throughout. Also each hand started with both players having a 20k stack. So each hand both players were equally deep.
2. This is just a guess, but I wouldn't be surprised if it would do reasonably well. The team behind Libratus has said that in their estimates Libratus could crush 6-max in 2 years. I think this is a careful estimate, and believe it can be done sooner if they put their effort on doing it.
3. Your guess is as good as mine here.
Computers can look at your hand thats why haha
Wayne Johnson the players are paired up and they get reversed hole cards to reduce variance. For example if Jason gets AA vs KK, Daniel will get KK vs AA.
oh thats very smart, they can't say they run bad
Well they can...the hands they played in poker terms is absolutely nothing...
Ultimate Bet style
Congratulations RUclips ai for finding me something interesting for a change!
I’m telling you these boys already had computers doing this online, I’m a great player, but when he said over bets and underbets, I get flashbacks to when these suspicious ids from sites would destroy me, I would always remember them cause they bet so random and just crazy , except it would work on me. There where only a few, but I’m sure I’ve played these bots that sites already have made.
Yup me too
Negreanu would still get a read
Four of the very best poker players that I have never heard of.. Have Phil Ivey play against the AI and then let's see how great that AI works
Not even watching the entire video because the title already sums it up. It should be "Can AI beat the best poker players in the world?", then i'd be watching the entire vid.
While I do kind of agree with you, the interesting part of the video was why (and how).
And this is how hydra started
The guy with the moustache just sounds like a robot! Weird cadence.
Ib4 the AI knew the opponents hand all along
Anyone else think the two scientists speak like robot text people use on RUclips videos?
Sweet video. Thank you.
1. War
2. Business
3. Disease...
Nice priorities, team.
I wish that it was available to the public like Poker Snowie!!
why didnt they have a hud
This was good. Can it win with a very finite amount of money to start with? I think that would be the real winner!
can anyone tell me how do i play against AI myself i want to try it is there any program/download i can use?
Guess we're making our own Gods now.
lol, he's giving a lecture there where the tables are empty, literally no players are playing around 7 am, I'm probably one of the weirdos playing lmao
That's pretty cool and all, but it doesn't take into account one of the most important aspects of poker - the fear of losing. The computer is not afraid to lose and so that has no effect on the decisions it makes, while for the human players it is a pretty significant factor, even at higher stakes. So in that sense, it has an unfair advantage over its opponents.
It does not work like that at pro lvl even though winning or losing still affects everyone at some lvl. What i want to say is that pros are less affected by this because they went through countless swings of variance along they pro life. They just simply play the solution they find and do not bother with the result. If they know they are better they know that eventually the EV line will converge.
In this situation they felt they are outplayed and not because of the result.
I guess I didn't express myself correctly when I said that fear has an even more significant impact at higher stakes. I agree with you that pro players are better at dealing with the pressure than regular players. That's one of the main reasons why they are more successful. However, even though they are less affected by the tension compared to normal people, they are still affected to some degree, especially when there are millions on the line. After all, they are still human and even the calmest, coolest person, sometimes can't help but feel a little doubt or insecurity under extreme pressure. I think that this is the main reason why there were outplayed - a human can never match a machine in this regard and so in a sense they were playing at a disadvantage from the get-go.
dan polk would have beat 'em
+onisuk If you don't get the joke, just let it go.
I was thinking that too. I was like wtf? Then I realized I wrote it. looooooooooolz
he already did
That was the old computer
0815hupe Doug polk
9:16 If I may add a slightly more grim approach.
(First of all, I want to say that I think tool development is one of the best things to help human kind overall)
But, the argument that he uses about having AI helping negotiations, by that point in time (and maybe, unfortunetly, it willl always happen), the "boss" will always have a better tool than the employee, specially for big corporations. Maybe progress will hault and become accesible, but we often have to think about the implications of our technological improvements.
i somehow believe overbetting turn card makes river plays much easier !
Ofc you need to be in possition/ALWAYS! If you're out of possition you should ALWAYS play trap game no matter if u AA or 72 holding.
These are not the best players in the world. They don't have much information about their success on the internet. They're playing 1500 hands a day which is an enormous amount, with 10 second shot clocks. Under those conditions, an OK professional player isn't going to be playing anywhere near the best. These players admitted to playing terrible strategies just hoping to find a weakness in the bot, rather than playing their own best strategies. The robot team is overstating their accomplishment here. I'd like to see this bot challenge the actual known best players in the world.
You are in denial. Hahaha
@@isaacvongurtberg7341 HAHAHAHHAHAHAHAhhHahHhHahahahhahahahahahHAHAAAAAAHHhahaaaHhaaaailoveyoudaddyHAHAHAHAHhhahahahahahHAHAHAHJAHAHAHHAHAHAHHAHAHAHAHHA
at the time this vid was made, these guys where probably in the top 10 of hu nl players. so yea, they were one of the best guys around
Anyone know what the ai looks like in a hud, like vpip 3 bet % etc
It is really weird when that scientist keeps calling the players 'humans'.
I still think a lineup of Dong Kim, Doug Polk, Daniel Cates, Ben Sulsky would have beat them. These guys were top 20 players but not the best other than Kim
I believe that Dong was considered the number two heads up no limit player in the world and he still had a slight loss against Libratus. Doug Polk is the only one who would've had a chance to beat Libratus.
so if i use ai to play online poker match, then i can win alot of money while im sleeping?
I do believe there's a precise way of kicking the AI butt.... 10000% randomness ..... everything all over the map, so it can't figure out your strategies the same way you can't figure out its .... much closer result i bet.
Have AI play against the worst poker players no chance. AI be like i cant figure him out
I'll kiss our future AI overlord's shiny metal ass.