Bob Mottram wrote:
2008/5/5 Richard Loosemore <[EMAIL PROTECTED]>:
 "The goal of symbol grounding is not to guarantee uniqueness but to ensure
that the connection between the symbols and the objects they are
systematically interpretable as being about does not depend exclusively on
an interpretation projected onto the symbols by an interpreter outside the
system."

 The crucial part is to guarantee that the meaning of the symbols does not
depend on interpreter-applied meanings.

Interpretors are constantly superimposing their own meanings onto the
raw data they observe.  There can be many different interpretations of
the same news story, or novel, or painting, or AI conference.  The way
I prefer to think about it is a continuous process of synchronising
predicted models with actual observations, because this encourages the
idea of perception and cognition as closed loop systems rather than a
tyranny of "bottom up" or "top down".

Personally I don't believe that symbols exist independently from
subjective interpretation (or a collection of such interpretations
commonly used within a society).  A spade is only a spade by virtue of
the way humans use it and tend to agree that it is the same kind of
object, otherwise it's just a peculiar configuration of atoms.

Symbols within biological systems are also far more amorphous than
their computational counterparts, existing only as groups of
synchronously active neurons (coalitions of the willing).

Um, well, I agree that symbols are in a sense subjective... but there may be a bit of confusion about my above quote.

I was pointing out that the 'interpreter' (i.e. the programmer) could build mechanisms that are only meaningful of the symbols conform to their interpretation of what the symbols mean.

But if the system itself then builds symbols and uses them in such a way that those symbols conform to the *system's* idea of what the symbols mean, everything gets seriously fubared if the two sets of interpretations don't agree. (which, in general, they will not).



Richard Loosemore

-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription: 
http://www.listbox.com/member/?member_id=8660244&id_secret=101455710-f059c4
Powered by Listbox: http://www.listbox.com

Reply via email to