[extropy-chat] Not necessary smarter, just faster?

[kevinfreels.com]

One problem. You would have to have an equally large 1 million fold increase in the speed of the person's actions. Working the brain a million times faster does not help if the eyes can't move and refocus fast enough to keep up, if they can;t communicate faster, and can;t get the actual physical part of the "work" done a million times faster. Wouldn't you just have a brain endlessly cycling while waiting for work to be completed.

----- Original Message ----- 
  From: Samantha Atkins 
  To: ExI chat list 
  Sent: Thursday, May 11, 2006 1:11 PM
  Subject: Re: [extropy-chat] Not necessary smarter, just faster?


  These points seem to be missing something.  A human level brain running at, say, 140 IQ but a million times faster can accomplish 1 million man years of work per year and do so without the tremendous management hassles and interpersonal friction of running a million person team.  That is huge.  Many problems are quite tractable to a large scale effort of that kind.


  It also seems  very likely that the neocortex would optimize many problems faster and more fully when run at vastly higher speeds with equivalently speeded up inputs.  Such a brain would be smarter over time and in much shorter time than otherwise.   


  There were no unique neural structures found in Einstein's brain AFAIK.  


  - samantha


  On May 11, 2006, at 3:54 AM, Robert Bradbury wrote:




    On 5/11/06, Eugen Leitl <eugen at leitl.org> wrote:

      You can compensate a lot by hard work. Up to a point. Vide supra: the eternal canine on fast-forward won't produce much than lots of happy barking, virtually gnawn bones, and tail-chasing.

    The problem with the "fast" canine example may be that the canine brain may not have either the (a) the capacity; or (b) the proper internal neural sub-nets to ever perform the function Einstein's brain did (recognition of some rather unusual laws of physics).  They might however have the internal subnets to extract information from smell data which humans completely lack.  (Say for example the "claimed" ability to be able to identify people who have cancer (or some types of cancer) based on smell.) 

    Running a neural network faster doesn't make it "better" at least for some things... A human brain on fast forward may still have a problem doing what some precisely adapted neural nets (an octopus or squid with highly precise sensory system processing and precision control of multiple arms) are capable of.  At the same time I don't believe those neural networks aren't particularly good at algebraic (symbolic) manipulation no matter how fast you run them.

    Einstein's brain may have had a unique neural structure so that it was able to make connections or recognize patterns that other brains simply could not (at least very easily).  Having (a) more memory capacity (human vs. a dog for example) or (b) better spatial manipulation capabilities ( e.g. those brains which can solve a Rubik's Cube [1] very quickly) or (c) better language sequencing capabilities (William Falkner comes to mind) may be things where faster does not equal more creative.  Though my general take on much "intelligence" right now is that similar brains (with ~ equal capacity and structure) can deal with almost anything given enough information, training and time.  Raw "speed" may help in getting from point A to point Z faster.  It is interesting to consider whether raw capacity (as compared to raw speed) is essential for solving the Professor's Cube [2].  This brings to mind space vs. speed trade offs in computer systems. 

    It raises the interesting question as to whether Einstein would have been able to deduce a "Theory of Everything" had his brain not been aging (over time brains do lose neurons) and/or had he been given another hundred or two hundred years to work on the problem? 

    Robert
    1. http://en.wikipedia.org/wiki/Rubik%27s_Cube
    2. http://en.wikipedia.org/wiki/Professor%27s_Cube 








    _______________________________________________
    extropy-chat mailing list
    extropy-chat at lists.extropy.org
    http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat




------------------------------------------------------------------------------


  _______________________________________________
  extropy-chat mailing list
  extropy-chat at lists.extropy.org
  http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20060511/a49a57a1/attachment.html>