Message-ID: <190320Z11071995@anon.penet.fi>
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!travelers.mail.cornell.edu!news.tc.cornell.edu!caen!zip.eecs.umich.edu!newshost.marcam.com!news.mathworks.com!gatech!news.sprintlink.net!EU.net!news.eunet.fi!anon.penet.fi
Newsgroups: comp.ai.shells
From: an306358@anon.penet.fi
X-Anonymously-To: comp.ai.shells
Organization: Anonymous forwarding service
Reply-To: an306358@anon.penet.fi
Date: Tue, 11 Jul 1995 18:54:23 UTC
Subject: The New "Cognitive Physics"p4of5
Lines: 60

The Rockford Analogy (part 4)

This is how an SSC self expands from one unit to a "cognitive task network".

Strictly speaking, a task network would be any two or more SSCs (my version
of an inductive learning machine) connected by their inherent expansion
bond(s). This bond would indicate the nature of the expansion.  The reverse
is also true, if one given unit expanded into two units, the type of
expansion would dictate the nature of the connection (bond) between these two
resulting units.  Also, units within an isolable task network will be related
by their common task. 

Practically speaking, a two-unit network would not be adequate to handle even
a comparatively simple decision environment.  Hence, expansion occurs.  In
default, or in an immature task network,  such expansion is primarily random,
not only as to the directions of expansion, but also as to the configuration
of the cognitive matter involved in the expansion event.

Normally, a fairly complex decision environment would typically induce the
generation of millions of "trial perspective" SSCs focused on an equal number
of unique configuration of cognitive matter.  However, only perhaps a
fraction (thousands) of these perspectives would be maintained (prove useful
in the target environment).

Stationary Observers in Cognitive Space

Cognitive space is a dimension in which a given single processing
instructions (or a cluster of defining particles) may be viewed as a tangible
isolable multi-dimensional artifact.  A sequence of such instructions, be it
liner or circular (a circuit), is an "assembly line" which conveys, creates,
combines, and alters such definition clusters.  A given perspective, from
which one views a given random configuration of artifacts, commonly presents
unique and useful profiles of the "Inherent Environmental Order" existing in,
 and emanating from, the task network's target environment.

(First Postulate of Cognitive Physics)
Once a processing structure has been created (even an imperfect one), one may
detect patterns reflecting useful Inherent Environmental Order (existing
within the nucleus-circuit's target environment) from any number of fixed
perspectives, involving any combination of the various isolable elements of
said processing structure (task network).

I am suggesting that by installing special "inductive learning machines" at
random "fixed perspectives" within the tangible dimension of the processing
machinery in cognitive space, one may solve (discover useful parallel order)
totally unknown (perhaps unsolvable by conventional reasoning) problems
simply by recognizing repetitive sequences and patterns (in the motion
[action] of these cognitive artifacts) from these fixed perspectives
(stationary observers) within said dimension of mechanized processing
structures.

(see part 5)
Copyright 1995.  All Rights Reserved.
by David Albert Harrell
DAHarrell@aol.com
----------------------------------------------------------------------------
To find out more about the anon service, send mail to help@anon.penet.fi.
If you reply to this message, your message WILL be *automatically* anonymized
and you are allocated an anon id. Read the help file to prevent this.
Please report any problems, inappropriate use etc. to admin@anon.penet.fi.
