BillK wrote:
Nobody has mentioned this yet.....
<http://www.physorg.com/news146319784.html>
I got a draft version of the paper earlier this year, and after a quick
scan I filed it under 'junk'.
I just read it through again, and the filing stays the same.
His basic premise is that connectionists argued from the very beginning
that they wanted to do things in a way that did not involve a central
executive. They wanted to see how much could be done by having large
numbers of autonomous units do things independently. Turns out, quite a
lot can be achieved that way.
But it seems that Asim Roy has fundamentally misunderstood the force and
the intent of that initial declaration by the connectionists. There was
a reason they said what they said: they wanted to get away from the old
symbol processing paradigm in which one thing happened at a time and
symbols were separated from the mechanisms that modified or used
symbols. The connectionists were not being dogmatic about "No
Controllers!", they just wanted to stop all power being vested in the
hands of a central executive ... and their motivation was from cognitive
science, not engineering or control theory.
Roy seems to be completely obsessed with the idea that they are wrong,
while at the same time not really understanding why they said it, and
not really having a concrete proposal (or account of empirical data) to
substitute for the connectionist ideas.
To tell the truth, I don't think there are many connectionists who are
so hell-bent on the idea of not having a central controller, that they
would not be open to an architecture that did have one (or several).
They just don't think it would be good to have central controllers in
charge of ALL the heavy lifting.
Roy's paper has the additional disadvantage of being utterly filled with
underlines and boldface. He shouts. Not good in something that is
supposed to be a scientific paper.
Sorry, but this is just junk.
Richard Loosemore
-------------------------------------------
agi
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
https://www.listbox.com/member/?member_id=8660244&id_secret=120640061-aded06
Powered by Listbox: http://www.listbox.com
