Talk:Computer poker player

Opening comment
This page is redundant with the pokerbots page.--Toms2866 14:19, 29 March 2006 (UTC)


 * Merged and redirected the orphan pokerbots page to this one which is referenced more, a bit broader in scope and older. 2005 20:42, 29 March 2006 (UTC)

I'm 95% sure that the mention of 'Combinatorial Game Theory' is incorrect. Probably what was meant was standard game theory (ie Von Neumann's stuff). —Preceding unsigned comment added by 80.229.247.11 (talk) 14:34, 28 October 2007 (UTC)

I agree, I think it's *not* "combinatorial game theory" but just plain "game theory". I changed it. 70.102.136.132 (talk) 23:20, 3 July 2008 (UTC)

Is poker computationally intractable? If so, need a cite. Smyjpmu 00:27, 9 November 2007 (UTC)

Tells
A major part of the skill of live poker games, however, is guessing at the strength of a player's hand by identifying tells made by other players, while concealing one's own. As a computer would not make any physical tells, playing against a computer would necessitate reading tells only from the bets placed; equally, the computer cannot read human tells.

I removed this because it just isn't true. Computers are not at all incapable of identifying tells. In fact, computer systems are currently being developed to detect lying by identifying facial movements. As far as I know this has never been adapted to poker, but not because it'd be impossible.


 * seems to me you could have reworded it and left it in. At the present time, AIs are not anywhere close to reading tells, but on the other hand, autonomous driving cars show that with enough data machine learning might figure it out. The article should reflect the truth about state of the art, not simply the theoretical limits. 199.83.222.137 (talk) 03:30, 6 November 2016 (UTC)

Cleanup tag removed
The article has now been cleaned up, the tag has been removed. Any queries, please contact me. Foxy Loxy Pounce! 08:47, 12 October 2008 (UTC)

Mirroring
I removed that section as it's utter crap. If poker were as easy as that I'd be a 25/50 regular by now; hand samples on which decisions are based in online poker are so utterly statistically insignificant that you have to have some basic strategy in most turn/river situations as it'll take several hundred hands to gain enough reads on your opponents thru sheer numbers (and obv he gets to extract info from showdowns while you only analyze inaccurate numbers). And that's for heads up play only, a "mimicking" strategy would be far more complicated at 6max or full ring play and the mirroring results of most bot programmers probably wouldn't even be able to beat other players, let alone rake —Preceding unsigned comment added by 93.136.201.234 (talk) 17:57, 29 January 2011 (UTC)
 * I found the page through your recent edits and I completely agree with you; it's crap. As a side-note, I strongly recommend getting a Wikipedia account! You have great insight and I feel like I could talk poker and stats with you for hours. JaeDyWolf ~ Baka-San (talk) 18:37, 29 January 2011 (UTC)
 * I agree. The whole article needs work as it is a whole lot of text without a single reference (despite having many external links). 2005 (talk) 00:29, 30 January 2011 (UTC)

Texas Hold'em solved
pgr94 (talk) 03:04, 9 January 2015 (UTC)

Artificial Intelligence
I removed what this section said about state space because it was definitely wrong, but someone that knows something about poker AI should rewrite this section. — Preceding unsigned comment added by Sadichen (talk • contribs) 06:30, 13 March 2016 (UTC)

PokerNews Citations
The IP address that keeps adding the citations from PokerNews might misunderstanding why these aren't legit citations. The main reason is that the first page is just a page shilling for GGPoker. Advertisements aren't reliable sources. The second page is arguably better, but the actual information is still limited to a single site, GGPoker, and the claims made are about "most, if not all," poker sites. It would be better to discuss this here than to continue to edit war. Rray (talk) 03:04, 14 February 2023 (UTC)