Re: Symbolic AI: Why Marvin Minsky and Curt Welch Are Out to Lunch

On 22 Jul 2006 09:24:13 GMT, curt@xxxxxxxx (Curt Welch) wrote:

Traveler <traveler@xxxxxxxxxx> wrote:
Curt Welch wrote:

Yeah, Traveler just posted the message saying the same thing - that I
don't know what a symbol is. You are both off your rocker. :)

When I speak a word, is that spoken word not a symbol? Spikes are not
the same as a symbol written on a piece of paper (and the most common
connotation of "symbol" is a fixed material object), but spikes are
exactly the same as spoken words and these do count as symbols.

The definition of a symbol is, "something that stands for or suggests
something else". When I say, "Bob is over there", does not the sound I
just made to form the word "Bob" act as a symbol of the person named
Bob?

I don't know where you guys learned what symbols are, but spikes are
definitely symbols. They are just short lived symbols that carry a
great deal of information in the temporal domain.

Nobody is accusing you of not knowing what a symbol is. From the
above, however, it's obvious that you don't know what is meant by
symbolic AI.

Why would that be since I wasn't talking about symbolic AI? Why is it that
all I have to do is use the word symbol and you think I'm talking about
what guys from the 50's were working on? Are you really that narrow
minded?

You wrote this in another thread (which led to this one):

For a computer like device to not be
able to duplicate the control functions performed by a human brain would
imply that humans were not able to understand the operation of the brain
(since our understanding is based on the manipulation of discrete symbols -
both words are our normal high level of existence, and pulses which are the
discrete symbols manipulated by the brain).

What could possibly be more symbolic AI than "pulses are the discrete
symbols manipulated by the brain"? Pulses are certainly not symbols
manipulated by the brain. Again, a symbol is representation, the
meaning of which is known a priori to the user, in this case, to the
brain. We all know this is not true (well, not all, since the symbolic
AI crowd are still championing this crap). That's the reason for the
pathetic failure of GOFAI. Fifty years of chasing wild geese and red
herrings. ahaha...

[cut]

I don't like long drawn out arguments. My position on this issue
should be clear to all at this point.

Louis Savain
.

Relevant Pages

  • Re: intelligence
    ... describe that that as a "code" for the control of the second signal. ... To me there are just pulses, ... brain as something very simple. ... That's consistent with my reinforcement learning approach to AI. ...
    (comp.ai.philosophy)
  • Re: intelligence
    ... describe that that as a "code" for the control of the second signal. ... To me there are just pulses, ... brain as something very simple. ... That's consistent with my reinforcement learning approach to AI. ...
    (comp.ai.philosophy)
  • Re: What evidence is there that a neuron is digital computer ?
    ... temporal resolution sensory input is pointless. ... But because pulses in the brain are actually analog, ... the practical temporal resolution limit of the neuron is ...
    (comp.ai.philosophy)
  • Re: Perceptual symbol systems
    ... So the mechanism of the brain could ... All they see are spikes and temporal patterns of ... Spikes and patterns of spikes should be the ... >> Hence the reason that GOFAI and its supporters are bankrupt. ...
    (sci.cognitive)
  • Re: Perceptual symbol systems
    ... >>behavior of a group of neurons as standing for some external stimuli ... In trying to understand brain function, ... All they see are spikes and temporal patterns of ... Spikes and patterns of spikes should be the ...
    (sci.cognitive)
Loading