You're definitely wrong on "No Limit will Never Be Cracked"
There is a Nash Equilibrium because it's a 2-player 0 sum game and, if anything, short stacked NL should be almost as easy as FL to solve because many of the decisions are push-fold and thus you can reduce the decision tree. Let's face it, Sub 10-BB Tournament poker is already near as makes no difference solved because Nash shoving ranges exist and have done for a while
In terms of deep stacked NL that's going to be a lot more complex but methods are constantly improving. 5 years ago everyone was beating the same drum that LHE would never be solved because there were too many variables so the UoA guys came up with a new way to make up for those variables in regret minimilisation and I imagine they could do something very similar to NL. Bet-sizing isn't insurmountable as many of the Poker AI competitors have shown you can defend vs many different bets sizes with just a few.
I wouldn't be surprised if you see 40bb- NLHE HUHU solved in the next 2-3 years and 100bb HUHU about 5 years or so after that and it'll be a long time before that where bots can beat the world's best players.
Polaris could beat the best in the world 8 years before Cepheus, the same will be true for NL bots
Additionally I'm not even sure Snowie's the best bot out there,according to the 2014 pokerAI competition that honour belongs to Tartanian7
"Team Name: Tartanian
Team Members: Noam Brown, Sam Ganzfried, Tuomas Sandholm
Affiliation: Carnegie Mellon University
Location: Pittsburgh, PA, USA
Non-dynamic Agent
Technique: One/Two Paragraph Summary of Technique:
Tartanian7 plays an approximate Nash equilibrium strategy that was computed on Pittsburgh's shared-memory supercomputer, which has a cache coherent Non-Uniform Memory Access (ccNUMA) architecture. We developed a new abstraction algorithm and a new equilibrium-finding algorithm that enabled us to perform a massive equilibrium computation on this architecture.
The abstraction algorithm first clusters public flop boards, assigning each cluster to a blade on the supercomputer. These public clusters are computed by clustering using a distance function based on how often our abstraction from last year grouped hands together on the flop with different sets of public cards. Within each cluster, the algorithm then buckets the flop, turn, and river hands that are possible given one of the public flops in the cluster, using an imperfect-recall abstraction algorithm. We did not perform any abstraction for the preflop round.
Our equilibrium-finding algorithm is a modified version of external-sampling MCCFR. It samples one pair of preflop hands per iteration. For the postflop, each blade samples community cards from its public cluster and performs MCCFR in parallel. The samples are weighted to remove bias.
Our agent also uses a novel reverse mapping technique that compensates for the failure of CFR to fully converge and to possibly overfit the strategies to the abstraction."
Although that said it plays incredibly wierdly atm and it definitely does some things wrong (I think it checked back a sf at one point...wierd AIs) and it looks completely insane to someone who's learned the "human" way of play NLHE but it's definitely a first step and would probably easily kill small stakes NLHU in it's current state so we'll see.
you should jump on my tables and lets see whos the better one at the game for old ladies and little children.
