Canadian Researchers Made Unbeatable Texas Hold'em Poker Algorithm

If it was me who made this i would not told anyone and just hit online poker sites until the of time.

Computer scientists say they've created an algorithm that has essentially solved a version of Texas hold 'em, and it's guaranteed to beat every single puny human competitor in the long run.

Cepheus, as this poker-playing program is called, plays a virtually perfect game of heads-up limit hold'em. The variant is like the popular Texas hold 'em, except there are only two players and a fixed number of bet sizes and raises. That still leaves 3.16 × 1017 states in the game.

The sheer feat of a program that's essentially solved a type of poker has computer scientists in a tizzy. Unlike other games that have been completely solved such as checkers or Connect Four, where every past action is laid clear on the game board, poker is a game with imperfect information—namely, you don't know your opponent's cards.

Can You Beat This Virtually Unbeatable Poker Algorithm?

Too bad is just a 2 player game and not real Texas hold'em, perhaps they just said its for this variation and they are using the real version hitting the tables as i would .

Does this mark the end of online poker?

Is this the beginning of skynet?

You can play against it in a link in the source

I started to come up with a method of playing video poker. With just you against the computer, you can be certain of the basic odds of getting a pair of jacks or better.

I tried it at the casino and would have scored pretty well if I played more money, but I pulled out after awhile when I lost $50. I didn't want to lose too much as it was my birthday. Besides, I was there for the concert anyway, not to gamble.

a reply to: Indigent
First off its head up limit holdem that has easier to calculate set variables throw a couple pros and few loose degenerate gamblers and even in a limit game the computer will lose 40 to 60 percent of the time which is better then the degenerates but not close to a pro who will win 80 to 90 %

Sure, the computer may beat me at poker,but I will dominate it at kick boxing.

.

originally posted by: Beartracker16
Sure, the computer may beat me at poker,but I will dominate it at kick boxing.

.

The computer will accuse you of kicking like a girl, and then quickly hack into and empty your bank account.

a reply to: Indigent

Curious if this is no limit or limit hold em....as that could be a huge difference. And with two players it sounds to me its only playing the odds which would become real predictable. But with 3 or more people you have to factor in things like pot odds...is going for that flush worth it with a 36% chance of making it? If 6 people are in the pot and then yes it is! 1v1 isnt that impressive...its just playing the odds. But interesting to see AI pushed forward in some realms.

Ha ha Ha Ha I think you know my computer.
Sometimes, I think it's out to get me.
a reply to: Aleister

a reply to: Indigent

Fixed sizes of bets and raises - That is what gets you in the end. All it does is let the program catalog a players pattern and next time he goes for it even unconciously the computer will anticipate it and better/play accordingly. There is no great mystery here just the fact that a program is more aware of your actions than you are and keeps track always.

The "algorithm" in this program is a calculation program based on mathematical probability?

a reply to: Indigent

a reply to: Asynchrony

the program is based on counterfactual regret minimization

if it helps

counterfactual regret minimization (CFR) (35).
CFR is an iterative method for approximating
a Nash equilibrium of an extensive-form game
through the process of repeated self-play between
two regret-minimizing algorithms (19,36). Regret
is the loss in utility an algorithm suffers for not
having selected the single best deterministic
strategy, which can only be known in hindsight.
A regret-minimizing algorithm is one that guar-
antees that its regret grows sublinearly over
time, and so eventually achieves the same utility
as the best deterministic strategy. The key in-
sight of CFR is that instead of storing and
minimizing regret for the exponential number
of deterministic strategies, CFR stores and min-
imizes a modified regret for each information
set and subsequent action, which can be used
to form an upper bound on the regret for any
deterministic strategy. An approximate Nash
equilibrium is retrieved by averaging each play-
er’s strategies over all of the iterations, and the
approximation improves as the number of itera-
tions increases.

this little puppy is what it takes to solve a 2 player poker game

Our CFR+ implementation was executed on a
cluster of 200 computation nodes each with 24
2.1-GHz AMD cores, 32 GB of RAM, and a 1-TB
local disk. We divided the game into 110,565
subgames (partitioned according to preflop betting,
flop cards, and flop betting). The subgames
were split among 199 worker nodes, with one
parent node responsible for the initial portion
of the game tree. The worker nodes performed
their updates in parallel, passing values back to
the parent node for it to perform its update, taking
61 min on average to complete one iteration. The
computation was then run for 1579 iterations,
taking 68.5 days, and using a total of 900 core-
years of computation

Being a poker player and having read tons on that subject. I can safely say that while technically there is an unbeatable way of playing poker (regardless of whether it is heads-up or 6 or 9 handed), this way is definitely not
the most profitable way and would pale in comparison in terms of profits to what you call playing exploitable poker.

This is because the unbeatable method of playing poker is a defensive scheme. One thing to take note is that human players can never play 100% perfect poker. Yes, skilled players can approximate playing optimal poker but it is just too complex to play it perfectly and many players would play with huge mistakes in reality. The unbeatable method is a defensive one as it is helping you NOT TO LOSE money, thus you gain money from opponent's mistakes. But, the exploitable method means playing in a way that it is not optimal theoretically so as to earn much more profits from a player who is playing with huge leaks in their game. High stake pros often use the exploitable method to earn as much from a fish (i.e a fish is a very bad losing player), when another pro notices and then tries to exploit the first pro (As the exploitable method is not unbeatable thus it can be exploited itself), the first pro would then revert back to a defensive method and one that is not exploitable or close too.

One way to compare it to the layman is imagine yourself playing Blackjack. You have studied the optimal way of playing Blackjack so you know exactly when to hit, when to fold, when to double/split etc. (But in this case if you discount card counting you are slowly losing money to the house playing optimally instead of slowing winning like in poker). Well one day, you are in luck! You notice that the cards are new and unshuffled thus they come out predictably like how it should from an unused deck. With this knowledge why should you play the theoretically optimal way when you can play an exploitable way that makes use of this knowledge to earn way MUCH MORE? This pretty much sums out why a technically unbeatable poker method is not the best method (And no humans are capable of playing it only close too). Also, you can forget about using bots as these poker sites would be able to detect and then kick you out.

originally posted by: MichiganSwampBuck
I started to come up with a method of playing video poker. With just you against the computer, you can be certain of the basic odds of getting a pair of jacks or better.

I tried it at the casino and would have scored pretty well if I played more money, but I pulled out after awhile when I lost $50. I didn't want to lose too much as it was my birthday. Besides, I was there for the concert anyway, not to gamble.

I don't think people got you but I did! Have a star for subtlety!

originally posted by: Indigent
If it was me who made this i would not told anyone and just hit online poker sites until the of time.

Computer scientists say they've created an algorithm that has essentially solved a version of Texas hold 'em, and it's guaranteed to beat every single puny human competitor in the long run.

Cepheus, as this poker-playing program is called, plays a virtually perfect game of heads-up limit hold'em. The variant is like the popular Texas hold 'em, except there are only two players and a fixed number of bet sizes and raises. That still leaves 3.16 × 1017 states in the game.

The sheer feat of a program that's essentially solved a type of poker has computer scientists in a tizzy. Unlike other games that have been completely solved such as checkers or Connect Four, where every past action is laid clear on the game board, poker is a game with imperfect information—namely, you don't know your opponent's cards.

Can You Beat This Virtually Unbeatable Poker Algorithm?

Too bad is just a 2 player game and not real Texas hold'em, perhaps they just said its for this variation and they are using the real version hitting the tables as i would

.

Does this mark the end of online poker?

Is this the beginning of skynet?

You can play against it in a link in the source

Rain Man and his fellow autistic and computer aided cheats have been banned from casinos for ever

If you play a straight mathematical game you may increase the odds in your favour which is why Casinos make mega bucks, they play full time 24/7 so will always win. But you cannot play 24/7 forever only game by game and you cannot guarantee the odds will stack in your favour in each game you play.