TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker

102 点作者 maurycy超过 8 年前

8 条评论

MikeTV超过 8 年前
&gt; DeepStack becomes the first computer program to beat professional poker players in heads-up no-limit Texas hold&#x27;em<p>Whether any others have been made before now is anyone&#x27;s guess. Botting is a known problem in online poker. If there&#x27;s a golden goose out there, I&#x27;m sure it&#x27;s being kept under wraps.
评论 #13359144 未加载
评论 #13361244 未加载
osti超过 8 年前
To be fair, none of the so called pros are considered big names in today&#x27;s no limit heads-up games. They should probably challenge ppl like WCGRider, Jungleman etc. next.<p>On another point, CMU just can&#x27;t seem to catch a break, their thunder continuously being stolen by UofAlberta in poker research, first in limit, now no limit. UofA clearly tried to publish this before the CMU poker challenge that&#x27;s supposed to begin soon.<p>To read more about the CMU challenge <a href="http:&#x2F;&#x2F;www.cmu.edu&#x2F;news&#x2F;stories&#x2F;archives&#x2F;2017&#x2F;january&#x2F;poker-pros-vs-AI.html" rel="nofollow">http:&#x2F;&#x2F;www.cmu.edu&#x2F;news&#x2F;stories&#x2F;archives&#x2F;2017&#x2F;january&#x2F;poker-...</a>
评论 #13359349 未加载
natecarroll超过 8 年前
The players they recruited were incentivized by a $8,000 prize pool up for grabs among the 34 of them...average $EV $235. They have to play 3000 hands to get a shot at that money, which is probably around 10 hours of multitabling. So that&#x27;s ~$24&#x2F;hr in expectation.<p>And then of course you don&#x27;t get anything unless you&#x27;re one of the top three winners against the bot, so there&#x27;s likely nothing to be gained from grinding out a marginal victory. You should just go ahead and play kinda stupid&#x2F;aggro and hope you win some of the big flips and whatnot. There&#x27;s literally nothing at stake for you except time value, so you might as well flame out early and then quit or run up a big stake to give yourself a shot at top 3.<p>Basically, the study design ensures the bot faces off against weak players playing in a way that would be sub-optimal in any other situation. Not surprised the bot won by a decent margin, nor that they are trying to spin this real hard in advance of the CMU poker bot matchup next week, which will be much more rigorous.
lawn超过 8 年前
I think I can wrap my head around neural nets being superior at games with perfect information like chess or go. But how would you teach bluffing and randomness to a neural net?
评论 #13358439 未加载
评论 #13358275 未加载
评论 #13358107 未加载
评论 #13358045 未加载
评论 #13358078 未加载
ChuckMcM超过 8 年前
I love it, research that pays for itself :-) I think of poker and other card games as imperfect but predictable information. So while you don&#x27;t know what cards the other players have you can certainly estimate the likelyhood of what they have and prune your choices that way. Think single deck card counting in Blackjack.
esseti超过 8 年前
the fact that they used hearts and spades instead of number for affilition is just lovely.
philosopheer超过 8 年前
most people (including here on HN) are complete n00bs when it comes to understanding how poker is played and how computers can play it, so just to straighten y&#x27;all out at the git-go here:<p>computers are <i>better</i> at bluffing and randomness than humans are. Bluffing is an important optimizing strategy in playing poker well, and it entails tracking the expected value of a pot (which includes cost expectations, don&#x27;t forget) and it entails randomness, necessary to obfuscate patterns of betting that could give away evidence of your bluffing strategy. Like chess and go, we may not be &quot;there&quot; yet with computers, but n00bs need to understand the theory.<p>What computers <i>can&#x27;t</i> do is read &quot;tells&quot;, so if you are a master poker player via tells (whether it&#x27;s unconscious or conscious thinking on your part) then you will beat other humans better than a computer will; but, by the same token, the computer will not give you tells to read nor be fooled by your fake tells. I think the mistake in thinking newbies (even highly experienced ones) make is mixing together &quot;the psychology&quot; of the game with the mathematics of the game.<p>So to give an oversimplified concrete example of a poker bluffing strategy (inspired by Nesmith Ankeny&#x27;s book), if odds of you drawing one of the cards you need to win a showdown are 1 out of 4 but the expected payoff is 20x then you not only need to stay in purely on expected value, but it is also an optimal time to bluff if you don&#x27;t get your card. It is informationally better to have a bluffing strategy that masquerades as an &quot;I have good cards&quot; strategy <i>and gives random information after the showdown</i> rather than &quot;bluffing&quot; being something you do sheerly when you have shit cards. And to enforce a <i>random</i> strategy on yourself, he recommends using a system of the cards in your hand as the random number generator to tell you whether to bluff or not: as you can see, his strategy designed for human players is more perfectly implemented by a computer.
评论 #13358370 未加载
评论 #13358320 未加载
评论 #13358666 未加载
brador超过 8 年前
Heads-up is solvable by just crunching the known probabilities, so i&#x27;m not sure what the achievement is here. Maybe the complexity of work involved to build the program is worthy of merit? Not sure.