Re: Neural netss (was Re: death of the mind.)

From: Glen Foy (spam33_at_butter.toast.net)
Date: 08/27/04

  • Next message: ZZBunker: "Re: Aaron Sloman's "The Irrelevance of Turing Machines to AI" article" 
    Date: Fri, 27 Aug 2004 19:57:56 -0400

    "dan michaels" <feedbackdroids@yahoo.com> wrote in message
    news:8d8494cf.0408271137.34cd8463@posting.google.com...
    > Wolf Kirchmeir <wwolfkir@sympatico.ca> wrote in message
    news:<_VGXc.30806$DG.1600198@news20.bellglobal.com>...
    >
    >
    > > The phrase may in fact mislead, since it implies a comparison (and the
    > > comparison has been explicitly invoked by some contributors to this
    > > thread.) But I don't see any comparison happening - there is AFAIK no
    > > "comparator module" that takes input from the "perception module" and
    > > the "face recognition module" and outputs "Yup, that's Johnny."
    >
    >
    > "... AFAIK ..." ... as far as you know. How can you possibly say there
    > is no compararator module, as far as you know? You're just making an
    > assumption.
    > ===============
    >
    >
    > The fact
    > > is that information about edges, etc is produced in the retinal layer,
    > > that this information passes to various NNs whose output goes to other
    > > NNs, etc until the final output is "That's Johnny." At each layer, the
    > > information is reorganised - that is, the inputs to the next layer have
    > > a different pattern. Note that the face recognition module is not as
    > > simple as the name impies - some people who have had damage in that area
    > > can still tell they are looking at a face, they just can't tell who it
    > > is. Others can't even tell they are looking at a face. Since the damage
    > > is to the NN, the "reference state" can only be the organisation of the
    > > NN, as I said above. Again, now what? The organisation of the NN
    > > determines how it operates, is all. Calling it a reference state doesn't
    > > help us understand how it's organised.
    > >
    >
    > In fact, given the anatomical organization of the visual cortex, with
    > its 30+ visual areas and 1100+ feedforward and feedback pathways,
    > there are roughly 1100/2 = 550 pathways which can feed back
    > "comparison" information. That's what basic FB is - comparison between
    > previous output and current inputs/state/etc activities.
    >

    The problem here, I think, is with the word "comparison". As Mr. Kirchmeir
    suggests, neural nets are not in the business of comparing anything. They
    accept inputs, and based on their interconnections they produce an output.
    The output can be recognition of some sort, the value of a function, etc.

    What feedback does is to *prejudice* the lower level networks by supplying
    inputs (positive and negative) when a higher level concept is active. An
    example would be helpful.

    My favorite epigram is "When all you have is a hammer, everything looks like
    a nail." Meaning, of course, that the higher level concept of hammer somehow
    influences the lower level perceptual mechanism. The lower level perceptual
    mechanism in this example is in charge of recognizing various types of
    fasteners, nails, screws, pegs etc. The concept hammer, when it is active,
    supplies inputs to the fastener recognition network so that it is
    predisposed to see nails. But from the point of view of the network, it is
    just processing its input, oblivious to any kind of comparison.

    Interestingly, as more and more tool concepts are developed, the prejudicing
    effect of the hammer is weakened. All the other concepts supply their own
    inputs, diluting the effect of the hammer.

    -Glen

  • Next message: ZZBunker: "Re: Aaron Sloman's "The Irrelevance of Turing Machines to AI" article"
    • Quantcast