It's actually pretty simple conceptually; all that is required is to walk through the entire game tree, look at what cepheus does in every spot, consider the probabilities for each of his hands given our hands (is cepheus a guy? derail...), and choose the action that maximizes ev. With some clever engineering and a large cluster of computers, the maximum ev of a perfect opponent can be computed relatively quickly (at least for huhu lhe).

There is at least one paper out there that outlines some of the tricks to making this computation run fast... sorry I don't have it on me on the moment.

bachfan's got it.

The biggest trick is probably computing values for all hands in ~1000 operations instead of ~1,000,000 operations: if you did things in the obvious way, looking at the value for one hand requires looking at ~1000 opponent hands and there are ~1000 hands. Described in http://poker.cs.ualberta.ca/publicat...t_response.pdf

pies01, if you were wondering about the specific advantage for the small blind and big blind, here's what we have. If you played perfectly as the small blind you would lose 87.718 mbig-blind/hand (4.3859 big blinds/100 assuming I can do basic math) to Cepheus. Playing perfectly as the big blind, you would win 89.691 mbig-blind/hand. Add that together, you would be ahead by 1.9727mbig-blind/2hands = 0.9863mbig-blind/hand.

(Just to make it clear, in this case by played perfectly I mean that you would pick actions which take the most possible money from Cepheus. You wouldn't have to play GTO to take Cepheus' money.)

Sorry, I meant epsilon equilibrium distance.
What I really mean is do you have EV data from specific nodes in the tree (as defined in CardRunners EV - which is what I think you mean by a public state tree). For example do you have the BB/SB EV at the start of the flop when the preflop action goes raise,raise,call, the flop then comes As7h9d and both players play according to cepheus. Can you tell us the EV for both players and the exploitability for both players from that node in the tree?

big blind has advantage vs cepheus?

Argh!!! Had to wait 90 minutes then the damn game froze and crashed on me after 10 hands.

Was down 1 SB at the time but I was likely ahead on the hand that play crashed. Only logical explanation is Cepheus had calculated that he was gonna lose so he froze me out.

So DOS attacks might be included in the GTO strategy?

how about a nice game of chess?

There's no such thing as a nice game of chess. It's a blood sport.

In the article, I believe they did mention application for diabetic research leading to self calibrating devices with self teaching.. something alongs those lines. so dont be quick to discredit them.

On a side note... Cepheus... Me and you, Flagpole 3PM buddy. Bring your BR, mines a mr.noodle and some mike and ikes.. and no I am not in jail.


rofl

Great stuff!





I wonder where would plo stand in that list.

PLO can be solved using precisely the same "brute force" methods as NLH. There's more combos to enter into the database, but the process is exactly the same. In the TP podcast, it sounded like the Alberta team has no interest in trying to solve PLO, because it's basically the same game as NLH, so finding a solution has no "real world" use. i.e. Once you've developed an algorithm for solving a poker game, there's barely any point in laboriously calculating the solution for each and every variant of that game. Altering the algorithms in order to treat diabetes or other diseases has a higher priority.

PLO play between humans is going to be much more about "feel" though. Since with close equity runs and huge hand ranges many decisions are gonna be indifferent. So there will be a huge number of mixed strats where you are supposed to bet from 1% of the time to 99% of the time etc.

Also there are things you can do in backgammon/chess to trick the computer (though admittedly a non-gto computer) and get a theoretically provable edge, but I'm not sure if you can do that in poker.

Still haven't been able to play this ****ing thing. I bookmarked the site and check it daily but it's always down for updates or says 'too many in queue'

Very well done sir. You spanked the bot in hand. Just for the record I did 2!

It's fantastic to get my run good out of the way in
a play money game.

You have to wait for a match to end and then hit the queue pretty fast. Took a little over an hour on wait list before my turn came up.

Not according to Joshua, the computer in War Games.

"Who about a nice game of chess.?"

Anyone posting results?

First attempt I won by 470. Gotta love rakeless games.

Won by 85, not going to lie I got tilted playing a computer

http://gyazo.com/b5317a37201212d629f81770fc15d55f

If this is just a big lookup table why can't this computer play hundreds of people at the same time??

Because it's not just a big lookup table.

I think both Joshua and HAL can beat this program.


........
They looked very tough in War Games and 2001: A Space Odyssey.