Page 1 of 1

Fast demonstrations of complex behavior

Posted: Sun Apr 01, 2001 2:09 am
by "Richard G. Dudley"
In response to comments by

George Richardson:
> ... I cant see what this teaches me about complex systems...

and Niall Palfreyman..

____________________

It would seem that the "Fast demonstrations of complex behavior" is a fast
demonstration of a simple agent based system. Each agent has a similar
program (remain equidistant from two other agents ).

I thought it provides a reasonable _example_ of an agent based system. An
emergent property, the location of the participants, is determined by each
person acting independently according to their programming.

A lightly more complicated example is provided when we take a bunch of
children onto a soccer (football) field. Half are dressed in red (red
agents) half in green (green agents). Plus a white ball. And two goals.
Each set of agents is given a set of directions (e.g. "kick the ball only to
similarly colored agents, or into the opponents goal").

If you already know what agent based systems are, I guess you dont learn
much. But for those who never thought about such things, it could be
enlightening.

_______________
Richard G. Dudley
rdudley@indo.net.id <mailto:rdudley@indo.net.id>
http://home.indo.net.id/~rdudley
http://ourworld.compuserve.com/homepages/drrdudley

Fast Demonstrations of Complex Behavior

Posted: Sun Apr 15, 2001 3:29 pm
by "Ian Fenty"
If you want to see that game that everyones talking about with the agents
that follow the simple rules and demonstrate complex behavior:
www.icosystem.net/Game.html
Peace,
Ian Fenty
From: "Ian Fenty" <ifenty@wpi.edu>

Fast demonstrations of complex behavior

Posted: Thu May 03, 2001 10:25 am
by Khalid Saeed
Bob,

I always build a simple model of the game and experiment with it with class
help when I debrief the game. In fact this debriefing is very helpful in
laying out the key threads of the modeling practice. Surprisingly, going
this route is also quite interesting and entertaining.

I have also found that SD modeling exercises in beginner level courses
should include substantial hand holding in terms of technical help in terms
of formulating the problem from description, building the model, and
setting up simulation experiments, while the student concentrates on the
interpretation of the experimental results. The students often continue to
implement the process we walk them through initially as we progressively
reduce this hand-holding in more advanced courses. Of course, we have the
luxury of having four years in our undergrad program to carefully bring the
student to a level of confidence where he/she feels comfortable with using
the modeling process they have learnt to address new problems.

Khalid Saeed
saeed@WPI.edu

_____________________________________
Khalid Saeed
Professor and Department Head
Social Science and Policy Studies
W. P. I., 100 Institute Road
Worcester, MA 01609, USA

Ph: 508-831-5563; fax: 508-831-5896
email: saeed@wpi.edu

SSPS Dept: http://www.wpi.edu/+SSPS