After alpha go beat the top go players, now top poker players cannot handle AI either. Data in a the future would destroy humans at poker. Soon all the battle filed commands would be generated by the chips and we are not going to be too far from terminator, not to mention all the jobs going to the chips today. https://www.pokernews.com/news/2017/01/poker-ai-beats-the-pros-26990.htm The poker pros took a bit of a beating during the “Brains vs. Artificial Intelligence: Upping the Ante” event that finished Jan. 30. The marathon 20-day poker competition had historic implications, with the artificial intelligence besting the pros. After the last of the 120,000 hands of heads-up no limit Texas hold’em was played, Libratus, the famed new computer from Carnegie Mellon University, had a collective $1,766,250 in chips, according to Carnegie Mellon. This is a statistically significant finding, suggesting that artificial intelligence can spread into other realms of information gathering, whether business negotiation, military strategy, cybersecurity or medical treatment planning. “The best AI's ability to do strategic reasoning with imperfect information has now surpassed that of the best humans," Tuomas Sandholm said, a co-developer of Libratus and professor of computer science with Noam Brown, a Ph.D. student in computer science, in a release from Carnegie Mellon. Having a computer that can communicate has widespread implications, explained Frank Pfenning, head of the computer science department at Carnegie Mellon. “The computer can't win at poker if it can't bluff,” Pfenning said in a release from Carnegie Mellon. “Developing an AI that can do that successfully is a tremendous step forward scientifically and has numerous applications. Imagine that your smartphone will someday be able to negotiate the best price on a new car for you. That's just the beginning.” Poker pros Dong Kim, Jimmy Chou, Daniel McAulay and Jason Les will split a $200,000 prize purse based on how they did in the event. McAulay and Les agreed that Libratus was tougher than they expected. “Whenever you play a top player, you learn from it,” McAulay said in the release. The AI computed its strategy on the Pittsburgh Supercomputing Center’s Bridges computer from Rivers Casino. That’s how the AI computed its strategy and played the four players at one time. Libratus, in contrast with other algorithms that calculate opponent weaknesses, used an algorithm to determine its own weaknesses from the other pros. Libratus’ end-game strategy also had a perfect analysis of the cards, according to Sandholm, updating its strategy during play. The computing power of Libratus, thanks to the Bridges computer, allowed it to solve an imperfect information game. “We designed Bridges to converge high-performance computing and artificial intelligence,” said Nick Nystrom, Pittsburgh Supercomputing Center’s senior director of research and principal investigator for the National Science Foundation-funded Bridges system, said to Carnegie Mellon. “Libratus' win is an important milestone toward developing AIs to address complex, real-world problems.” Be sure to complete your PokerNews experience by checking out an overview of our mobile and tablet apps here. Stay on top of the poker world from your phone with our mobile iOS and Android app, or fire up our iPad app on your tablet. You can also update your own chip counts from poker tournaments around the world with MyStack on both Android and iOS. M
I'm honestly surprised. I know it immediately knows all the odds, but not much better than the best poker players. And then it can't really *read* the table, right? In terms of watching people's faces, etc. But maybe that's a wash, because nobody can read the computer's "face" either. Really remarkable and maybe scary milestone. Like, if a computer can bluff, what does it do with its spare time when you walk away from it and the poker game is over. LOL.
At this rate self aware AI is coming in the future for sure I would say, how the human race deal with that is hard to say. Would we have Rosie or Terminator in the future? Maybe both, Transformers vs Decepticons?
There's been a very interesting AMA with the inventors and players on Reddit, it answers most of the questions.
I followed this pretty closely and watched the Twitch feed pretty regularly during the tournament. I'm the top question asked on the AMA lol. The bot is a freaking beast. Perfectly balanced, impossible to read.
Most of "reading" a player at the poker table has nothing to do with their expressions and nearly everything to do with their betting patterns. This is data that a computer can easily analyze. ETA: This is true for poker pros. Amateurs can be "read" by their expressions much more easily.
True, yes. I still have always thought, however, that really perceptive social mammals get implicit nonverbals (and maybe even subliminal chemical) signals from other mammals. I've played a lot of poker and still think there's at least a *little* something beyond reading betting patterns when it comes to reading a table. But then, I don't play at the high levels where people have ironed out a lot of their readable reactions, etc, etc.
Try tables where they are actively trying to use cues to distract or mislead. That's where it gets really hairy.
Futurists are pushing this leap as a form of augmentation rather than a full replacement. The car salesman v. mobile phone example mentioned before would mean that two AI duke it out for your cheaper price. It's pretty much abstracting away the human skill of negotiation much like how we've abstracted aerial combat to the machines. Machine learning is still a big black box. The bright side of it is that because it's still new, people can pick it up with a fairly decent gaming rig. Or spin one up on a cloud server and watch it burn away money and electrons.
With the advancement of AI, 9 out of ten career currently in existence can be made obsolete if you put enough capital into it. So in the future the top .1% will control all the AI robots including the army and police AI, the rest of us will live in slums. Or we all party and the AIs do all the work until they decide to get rid of us.
I like kooky Ray Kurzweil's take on it. Is it that far a stretch to have cellphones in our bodies? Instead of Skynet, maybe it's a Robocop situation that we should be worried about.
Maybe I don't understand this stuff. Looks impressive, but it's still human applied alg that is being used. An awesome tool to accomplish a set of specific tasks. What's impressive is if a machine can pick up that skill over time with self-learning. I guess you call that machine learning.
Check out alpha go, the new version is going to learn how to play go purely from self learning, think of what the future will hold, AI directing traffic in the air and on the ground, AI guiding and maybe launching missiles an bombs, AI doing business negotiations, the possibilities are endless, unless you are the ones being replaced.
Cool. Come to think of it. What is machine tool but to specialize in an area. Machine can already do much better job than human in many areas and these additional "intelligent" abilities, is an expansion of that into areas that are dynamic. A self driving car. A poker bot. Military autonomous drone. Machine language and deep learning is another leap, or side leap, but I believe still specialize in an area (learn one or a few specific functions or data model - again, self driven car; speech recognition and response). When a machine can self learn generally across multiple areas - it is closer to human, but that isn't what we want (I don't think). We want tools, helpers, enhancers to humans. What I don't have a clue is can these machine make that "self" leap from being specialized to generalize (human intelligent), or it really does not need to make that leap to endanger humans - an out of control set of military drones ... ?? p.s. I'm always interested in statement that machine will take over human... can be dangerous... is it remotely true? Ah, just blabbing here.
Many of the top experts in those fields do think so, I think the claim is indeed valid and concerning. One issue is that the top AIs are capable of self-learning and even the developers don't fully understand what is going on in their system, other experts have also indicated that machines which are based on efficiency will develop drives to ensure they always reach their goal in the most efficient way, which could lead to humans being seen as obstacles.
