Need ideas about building smarter agents

[pauljohn]

I wonder if anywone has the El Faroul simulation written for swarm?

I'd like to build a model where agents can learn and adapt their
strategies, and I'd really like some examples how to do it. Didn't
someone in here talk about a stock market model with agents that learn
(not like the one distributed with swarmapps?).

On another note:

I made a tarball of a simple swarm program called
 "MajorityRule-1.1.tgz". It might be useful for some teaching purposes,
as it is simple--rather like simpleSwarmBug.  There are voters who
have "ideal points" in 2D, and a single agent-- "the agenda setter"--
moves policy by making proposals (according to distributional
assumptions) to win majority support.  It shows the movement of policy
in a 2D continuous space according to the will of the majority.
Interesting differences are found by changing the width of the angle
that governs the policy proposal process as well as the number of
agents. 

This might be good for teaching because 1) its not about heatbugs, 2) it
deals with some problems of dimensionality that new users will confront,
3) deals with the problem that most swarm code requires type id as
input but the model was not originally suitable for that (hence the
new type "dimarray" that gets wrapped in an object so agents can pass
it around among themselves.  It has plenty of stuff to tinker with. 
There is a fair amount of commenting in the source.  I've never
distributed code before, so 
I apologize in advance for the difficulties users might have. This
steals code examples from heatbugs and some of the other SFI freebies.
Also special thanks to Glen R and Benedikt S for their patient
counseling while I was learning about this.


Do I just ftp this into the incoming directory at SFI?  Until I
understand how I'm supposed to put it there, I'm going to put it in my
homepage Swarm directory.  If it doesn't work, please let me know.

http://lark.cc.ukans.edu/~pauljohn/Swarm

-- 
Paul E. Johnson                       email: address@hidden
Dept. of Political Science            http://lark.cc.ukans.edu/~pauljohn
University of Kansas                  Office: (913) 864-9086
Lawrence, Kansas 66045                FAX: (913) 864-5700


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.
                  ==================================