2012 (2): The House Always Wins
The most memorable conversation I had at the ESOMAR 3D conference this year wasn’t about research at all: it was about poker. Josh, from BrainJuicer’s marketing team, is a keen online poker player, and explained the extent to which online poker in particular is a percentage game – playing for incremental gains on multiple virtual tables. All the regular players, he said, are using apps and algorithms to automatically calculate their odds and are adjusting their play accordingly. Which leaves the newbies and the suckers who want to see poker as a social game, or imagine it’s a game allowing for flair, or employ a range of other losing strategies which are generally based on individuality rather than trusting the numbers.
Reminds me of this great article by Gary Kasparov that appeared in the NYRB a couple of years back. It’s about how good computer chess AIs have affected the game. The kicker, talking about ‘freestyle’ tournaments in which any combination of people and computers may play in teams:
The surprise came at the conclusion of the event. The winner was revealed to be not a grandmaster with a state-of-the-art PC but a pair of amateur American chess players using three computers at the same time. Their skill at manipulating and “coaching” their computers to look very deeply into positions effectively counteracted the superior chess understanding of their grandmaster opponents and the greater computational power of other participants. Weak human + machine + better process was superior to a strong computer alone and, more remarkably, superior to a strong human + machine + inferior process.
I suspect that the poker players mentioned in Tom’s post have similarly refined a process. Good poker AI is a hard thing but weak poker AI (essentially probability calculators with nice interfaces) + good human process + playing across multiple tables (smoothing out the noise) + drunk opponents (A guy I know who makes money from this kind of thing always plays sober on Friday nights)…
From the link above - why the game is interesting:
- hidden or imperfect information, a fundamental problem arising in many AI domains;
- deception, an important property that is critical to success;
- multiple players (two to ten), increasing the complexity of the domain;
- stochastic outcomes, introducing variance and making evaluation difficult; and
- variable payouts, forcing players to maximize their winnings rather than number of wins.
Richard ‘Magic the Gathering’ Garfield offers another perspective on why poker is interesting
-
afarrell reblogged this from blackbeardblog and added:
being someone with flair / individuality. More...more aggregated it gets - people don’t...
-
lonepilgrim reblogged this from tomewing and added:
my muddled mind, this seems like...Tom’s Tumblr article
-
tomewing liked this
-
toffeemilkshake reblogged this from blackbeardblog and added:
great article by Gary Kasparov that appeared...years back. It’s about how good computer...
-
jrichmanesq liked this
-
imathers liked this
-
nickdouglas liked this
-
tomewing reblogged this from blackbeardblog and added:
More end-of-year crystal/navelgazing - somewhat more dystopian...be honest I think...
-
blackbeardblog posted this
