Stories
Slash Boxes
Comments

SoylentNews is people

posted by Dopefish on Sunday February 23 2014, @12:00AM   Printer-friendly
from the knowledge-is-power dept.
dyslexic writes "An Equation For Intelligence? It is something like the philosopher's stone. A sort of E=mc2 that would put intelligence, and more particularly artificial intelligence, on a sound theoretical footing. But could it be as simple as this TED talk video (available on the link in addition to the article) suggests? The video explains some of this and provides examples of the principle in action where it is claimed to replicate a number of "human-like" intelligent behaviors including cooperation and tool use."
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 5, Funny) by unitron on Sunday February 23 2014, @12:09AM

    by unitron (70) on Sunday February 23 2014, @12:09AM (#5015) Journal

    Just cobble together something where intelligence exists in inverse proportion to a willingness to impose beta on your "audience".

    --
    something something Slashcott something something Beta something something
    • (Score: 0) by Anonymous Coward on Sunday February 23 2014, @11:49AM

      by Anonymous Coward on Sunday February 23 2014, @11:49AM (#5153)

      Wouldn't that be the equation of "dumb"?

  • (Score: 5, Insightful) by frojack on Sunday February 23 2014, @12:16AM

    by frojack (1554) on Sunday February 23 2014, @12:16AM (#5016) Journal

    Sorry for not playing along, but if you can't be bothered to summarize it I can't be bothers to actually address the issue.

    --
    No, you are mistaken. I've always had this sig.
    • (Score: 5, Insightful) by kebes on Sunday February 23 2014, @01:21AM

      by kebes (1505) on Sunday February 23 2014, @01:21AM (#5032)
      I watched the video. The speaker's style is a bit monotone, some of the connections he draws seem unfounded (even having a pseudo-scientific "I can explain everything!" flair about them), and his equation is ultimately not grounded in anything physical (so not really useful). So, I don't really recommend watching the video.

      Nevertheless, he has some interesting ideas; and what he's proposing is at least a testable hypothesis. At the end, he summarizes his hypothesis as:

      Intelligence should be viewed as a physical process that tries to maximize future freedom of action and avoid constraints in its own future.

      He gives various examples of how intelligent behaviour often involves taking actions that maximize future options: good game-playing strategies (for humans and computers) often involve maximizing options and avoiding boxing oneself in, intelligent species make longterm plans to avoid death (the ultimate lack of options), social connectivity can be thought of in terms of maximizing options, etc.

      I find it unlikely that this single idea will fully explain all aspects of intelligence. But it may well be a crucial component, or at least a valuable way to frame the problem.

      • (Score: 4, Insightful) by Ethanol-fueled on Sunday February 23 2014, @01:34AM

        by Ethanol-fueled (2792) on Sunday February 23 2014, @01:34AM (#5033) Homepage

        To add to your skepticism, it sounds like an oversimplification of an all-encompassing and nebulous subject from a monotonous nerd's perspective, referring to the narrow case of board games and using thermodynamics and information theory as inspiration. The irony is that he says that intelligence and maximizing options go hand-in-hand, while he is explaining using few options with his narrow point of view. What would you say about somebody who was intelligent but set in their ways and didn't like being presented with options?

        Kinda like Freud, who was one smart motherfucker but (in my opinion) had a tendency to project his own psyche a little too much into what was also a complex and nebulous subject.

         

        • (Score: 0) by Anonymous Coward on Sunday February 23 2014, @12:35PM

          by Anonymous Coward on Sunday February 23 2014, @12:35PM (#5165)

          Freud made fictitious case histories. There is no science to back up anything he ever wrote. It was an injustice the way that schizophrenia was blamed on bad mothering and women were made to feel guilty for their child's mental illness. You can't say that no one knew any better - biological psychiatrists certainly did know better for a long time. Freud was a charlatan.

    • (Score: 5, Insightful) by istartedi on Sunday February 23 2014, @01:51AM

      by istartedi (123) on Sunday February 23 2014, @01:51AM (#5036) Journal

      But, but... It's a TED talk! All the best
      people are there and the charts and graphs are
      so delicious to look at. Also, each and every
      talk revolutionizes the way you think and the
      world you live in. By the time you get out, you
      won't know what to think or where you are. It's
      the thinking man's weed.

      --
      Appended to the end of comments you post. Max: 120 chars.
      • (Score: 5, Insightful) by Jerry Smith on Sunday February 23 2014, @07:30AM

        by Jerry Smith (379) on Sunday February 23 2014, @07:30AM (#5110) Journal

        Also, each and every talk revolutionizes the way you think and the world you live in. By the time you get out, you won't know what to think or where you are.

        Oh man, I hear you. Several of my students are complete fans of these talks and BECAUSE it's a ted talk it must be true, otherwise it would not be a ted talk. They absolutely stop thinking for themselves once the video starts. Pointing out flaws or straight out lies are considered to 'spoil it'.
        I can't be bothered to watch them myself, and proper transcripts somehow aren't there. No, subtitles don't count.

        --
        All those moments will be lost in time, like tears in rain. Time to die.
    • (Score: 4, Insightful) by WildWombat on Sunday February 23 2014, @02:48AM

      by WildWombat (1428) on Sunday February 23 2014, @02:48AM (#5049)

      Its a rare occurrence when I watch a video of that type, summary or not. In general I feel that if you have an idea you want to convey, write an essay. I can read your essay in way less time than it will take to watch your video. If you really need to convey some visual information, use some pictures. I know everyone learns differently and that videos may be great for some people but I almost completely ignore video as a format for intellectual discussion. Its just inefficient time wise.

      Cheers,
      -ww

      • (Score: 4, Insightful) by FatPhil on Sunday February 23 2014, @11:18AM

        by FatPhil (863) <pc-soylentNO@SPAMasdf.fi> on Sunday February 23 2014, @11:18AM (#5148) Homepage

        Oh, I always watch TED talks that are linked to. Just to confirm my opinion that they're self-indulgent fatuous nonsense.

        This guy likens himself to Einstein and Feynmann.

        Best about that is that he likens the importance of his equation to Einstein's E=mc^2. He seems to be overlooking that there's no such equation. Einstein came up with E^2=(mc^2)^2+(pc)^2. The variable p term later got merged in with the constant m, and created a variable ("relativistic") m. Einstein was horrified by this modification, and denied its validity. Sure, the short equation is punchier, but why sacrifice correctness for space?

        So he's likening his equation to a misleading misattribution.

        I guess he did that to maximise his future choices...

        --
        Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
  • (Score: 4, Funny) by BsAtHome on Sunday February 23 2014, @12:18AM

    by BsAtHome (889) on Sunday February 23 2014, @12:18AM (#5017)

    I'm sorry, but that was answered a long time ago, took 7.5 million years to calculate and is 42. How difficult can that be?

  • (Score: 5, Funny) by andrew on Sunday February 23 2014, @12:28AM

    by andrew (755) on Sunday February 23 2014, @12:28AM (#5018)

    could it be as simple as this TED talk video

    Everything is simple in a TED talk. http://youtu.be/DkGMY63FF3Q [youtu.be]

    • (Score: 0) by Anonymous Coward on Sunday February 23 2014, @05:38AM

      by Anonymous Coward on Sunday February 23 2014, @05:38AM (#5089)

      So, the anwser is, no. Something about some law about headlines, that I once read about on /. that came from a TED talk. Sorry.

  • (Score: 5, Funny) by CluelessMoron on Sunday February 23 2014, @01:06AM

    by CluelessMoron (1374) on Sunday February 23 2014, @01:06AM (#5029)

    Ok, with "E = mc²" E is in Joules, m in kg and c is the speed of light in m/s, it works out in dimensional analysis, and actually matches experiments. The equation itself can be worked out from scratch on one sheet of paper.

    With this "F = T∇St" wtf are any of the units? Gibberish about "causal entropic force" and "notional temperature"?? Oh please. This is no better than A = B+M, which means "Awesomeness equals Beer plus Music".

    Stick with Time Cube please.

    (and holy shit, Unicode works on SN?)

    • (Score: 2) by mtrycz on Sunday February 23 2014, @02:40AM

      by mtrycz (60) on Sunday February 23 2014, @02:40AM (#5046)

      Unicode works on SN?

      Some of it, yes. On odd days of the third fase of the moon.

      --
      In capitalist America, ads view YOU!
  • (Score: 5, Insightful) by lubricus on Sunday February 23 2014, @03:55AM

    by lubricus (232) on Sunday February 23 2014, @03:55AM (#5064)

    OK, I watched your TedX talk.

    It seems that the equation is really just a nice metaphor, a useful toy model of behavior.

    As with all Ted talks, this one is overly simplified, and seems to maximize "coolness" over detail. this is the problem with Ted talks in general (What's the problem with TED talks) [youtube.com].

    Here's his own concluding statement:

    Intelligence should be viewed as a physical process that tries to maximize future freedom of action and avoid constraints in its own future.

    Allow me to make a parallel statement:

    Is there an equation to live a good life?

    A good life should be viewed as a deliberative process that tries to maximize happiness while avoiding bad outcomes.

    OK, that was a little trite, but you get my point. Sure it might be food for thought, and might be true in the most general sense, but it's also hopelessly vague, and skips over all the most interesting detail: how to determine if something maximizes happiness.

    Similarly, how one determines what "maximizes future freedom", or "avoid constraints" is the interesting, difficult part.

    It seems like he wrote a machine=learning algorithm which attempts a simple optimization, maximizing future entropy, saw that it "worked" for various systems, and perhaps took a leap to conclude that this is the basis for all intelligence.

    --
    ... sorry about the typos
    • (Score: 3, Interesting) by Anonymous Coward on Sunday February 23 2014, @04:15AM

      by Anonymous Coward on Sunday February 23 2014, @04:15AM (#5071)

      "It seems like he wrote a machine=learning algorithm which attempts a simple optimization, maximizing future entropy, saw that it "worked" for various systems, and perhaps took a leap to conclude that this is the basis for all intelligence."

      Yes, exactly. It seems his concept of what is intelligent is based on very human emotion or whim. His initial concept of deflecting asteroids in order to survive, and that being a core sign of intelligence, is entirely subjective, it could also be viewed as fear or desparation. Accepting death and watching the asteroid crash could be viewed as even higher form of thinking, if you wish. Similarly with his ball and stick balancing computer. Because humans think balancing a ball on a stick is clever, especially if we "think up" the idea given a ball and stick, he suggests the computer is smart doing it. Why hasn't it realised it is also futile and been a little more creative? And as for winning at the stock market, is this not just greed? He's just finding an equation to reenact some of our more flawed and self centered behaviour.

      • (Score: 1) by lx on Sunday February 23 2014, @06:12AM

        by lx (1915) on Sunday February 23 2014, @06:12AM (#5091)

        While I sympathise with your core argument, I don't think that giving up and slacking off are signs of great intelligence. A lack of (perceived) options is highly correlated with anxiety and depression in most mammals including humans.

        • (Score: 2, Insightful) by Anonymous Coward on Sunday February 23 2014, @12:00PM

          by Anonymous Coward on Sunday February 23 2014, @12:00PM (#5156)

          OK, but you're actually just proving my point. YOU don't think giving up is a sign of great intelligence. It's your opinion, others might disagree with you, therefore the premise for any equation built on that "belief" is as subjective as your particular view. Where does my view fit in? Or am I not intelligent....

          As for the stressed mammals and depression. In sure that's true, but depression isn't a sign of reduced intelligence, and as far as I could see, there is no happiness quotient built in to the equation whatsoever. And again it's entirely subjective, many humans and mammals have few "options", sometimes from choice. Some religions or mental exercises, like zen, produce a perceived higher consciousness through nihilism our acceptance of just one option. The asteroid just is.

  • (Score: 3, Interesting) by TGV on Sunday February 23 2014, @07:37AM

    by TGV (2838) on Sunday February 23 2014, @07:37AM (#5111)

    The relation between information and entropy is that they can be expressed similarly. But information is something in our heads, not in the state of a gas. Claiming that information and entropy are intertwined doesn't help understanding intelligence.

    But the real problem of the equation is in computing the entropy. How do you know which path is going to yield the best options? You'd still need an evaluation function for the states, and that's precisely where the intelligence is. Not in some steepest gradient search function.

  • (Score: 5, Informative) by doru on Sunday February 23 2014, @07:58AM

    by doru (319) on Sunday February 23 2014, @07:58AM (#5114) Homepage

    This is based on a paper [doi.org] published last year. The reprint [alexwg.org] is available from the author's site.

    • (Score: 1) by linsane on Sunday February 23 2014, @05:32PM

      by linsane (633) on Sunday February 23 2014, @05:32PM (#5238)

      An interesting related article in New Scientist relating to pain and how certain types of beastie have evolved to sense it in different ways, from 'not at all' (insects) through 'sort of' (squids) to 'most definitely a bit like humans' (octopi) http://www.newscientist.com/article/mg22129570.600 -d-invertebrates-feel-pain.html [newscientist.com]

      The link between pain and self awareness is one that I've never properly considered before, won't stop be eating whatever I feel like though :-)

  • (Score: 2, Funny) by weeds on Sunday February 23 2014, @04:18PM

    by weeds (611) on Sunday February 23 2014, @04:18PM (#5204) Journal

    "No"

    see: Betteridge's law of headlines http://en.wikipedia.org/wiki/Betteridge's_law_of_h eadlines [wikipedia.org]