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

From: Eray Ozkural exa (erayo_at_bilkent.edu.tr)
Date: 09/18/04 

Date: 17 Sep 2004 20:44:25 -0700

feedbackdroids@yahoo.com (dan michaels) wrote in message news:<8d8494cf.0409110908.dd8b48a@posting.google.com>...
> erayo@bilkent.edu.tr (Eray Ozkural exa) wrote in message news:<fa69ae35.0409110231.57dd9767@posting.google.com>...
>
>
> > > So: What low level languages have "emerged" in NNs of terrestrial organisms?
> > >
> > > And how can you tell?
> >
> > We would be able to reverse engineer these languages directly with a
> > sufficient theory and data, but a theoretical approach is better,
> > let's try to construct such languages, or evolve them on a computer...
>
>
> Hi Eray, you might be interested in taking a look at comparisons
> between computationalism and so-called dynamicism ....
>
> http://www.google.com/custom?&q=computationalism+dynamicism

Philosophy of computation encompasses continuous computation, which is
what the dynamicists are trying to say discrete computation lacks. You
don't have to be a genius to see that.

The neural networks are just a particular parallel architecture, for
what it accomplishes, nothing more, nothing less.

Computation, as a concept, is more general than "evolution of
dynamical systems", in my opinion. What is a static system? :) What is
it that evolves in a dynamical system? How does it evolve? What are
the limits of its change? All sensible questions that can be asked
about a dynamical system are computational. Ah, of course, you can
attach pretty words to that, as well.

BTW, why should we think that continuous functions should physically
exist? I don't see the physical evidence!

Regards, 

--
Eray Ozkural