Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!pipex!uunet!newsflash.concordia.ca!nstn.ns.ca!cs.dal.ca!ug!mitchell
From: mitchell@ug.cs.dal.ca (Dave Mitchell)
Subject: Re: Flocking?
Message-ID: <Cy4vKt.6HC@cs.dal.ca>
Sender: usenet@cs.dal.ca (USENET News)
Nntp-Posting-Host: ug.cs.dal.ca
Organization: Math, Stats & CS, Dalhousie University, Halifax, NS, Canada
X-Newsreader: TIN [version 1.2 PL2]
References: <Cy15B1.3p5@cs.dal.ca> <Cy18Dq.6Cr@cs.dal.ca> <38bs3v$8kd@news.acns.nwu.edu> <38c8b1$l8c@news.halcyon.com>
Date: Sun, 23 Oct 1994 16:10:04 GMT
Lines: 21

James K. Peckol (jkp@halcyon.halcyon.com) wrote:
: with both flocking and schooling, we have what we can view as a computer in 
: which the solution to a problem is given by the state of the system
: (similar to a neural net).  more over, the solution is computed by a
: collection of elementary computers, each operating with a simple instruction 
: set.  it is not immediately obvious if the individual components observe
: and react to the state of the system as a whole, to only the behavior of
: n nearest neighbors, or to a combination of both.  

: has anyone done any work in trying to apply such an idea to a computer
: architecture. i'm familiar with some of the early work with systolic
: arrays, and i think this is a different phenomena.  would be interested
: in hearing other ideas.

I wonder, isn't this about the way that a large parallel asynchronous
computer would work? I haven't done any architecture (but I read about it
in BYTE!) Of course I doubt if the individual processors in the machine
would intentionally exhibit variations in response as various people would.


							thanks-Dave
