swarm-support
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: classifier system routines


From: Rick Riolo
Subject: Re: classifier system routines
Date: Tue, 14 Jul 1998 22:13:03 -0400 (EDT)

Hi,
CFS-C certainly is not still state of the art, neither
implementation-wise nor conceptual-wise!
(After all, its over 12 years old...when I wrote it
the ansi-c standards were not yet in place, and various
compilers on pc's and mac's had strange limits on heap
sizes, variable name lengths, etc!)

We have implemented an alpha version CFS-S in Swarm.
You can get a version (which works under some earlier Swarm version,
eg 1.0 or so...I can't recall right now) 
at the PSCS website:
  http://www.pscs.umich.edu/Software/Contents.html

This implements all that we still thought might be useful
in CFS-C, plus some new features and with plans
to make it easy to implement some of the features
in other (non-public) systems, eg Stewart Wilson's
macro classifiers.

Please feel free to download this alpha version and give it
a try.  Feedback is welcome.   I hope to get someone
to work on it in the next year sometime, to
a) get it up to the latest swarm
b) find and squash bugs
c) implement new features (eg macro classifiers)

- r

ps By the way there are newer non-swarm classifier
   systems in the public domain, which I believe you can
   find at the various Encore archives, eg, two offspring
   of the original CFS-C, and Tolkien, a C++ based system.
   There might be others as well, I'm not sure.

Rick Riolo                           address@hidden
Program for Study of Complex Systems (PSCS)
4477 Randall Lab                
University of Michigan         Ann Arbor MI 48109-1120
Phone: 734 763 3323                  Fax: 734 763 9267
http://www.pscs.umich.edu/PEOPLE/rlr-home.html

On Tue, 14 Jul 1998, Pietro Terna wrote:

> Date: Tue, 14 Jul 1998 23:19:42 +0200
> From: Pietro Terna <address@hidden>
> To: address@hidden
> Cc: Gianluigi Ferraris <address@hidden>
> Subject: classifier system routines
> 
>       Hi all,
> 
>       with two of my students, I'm planning to embed Rick Riolo's CFS-C 
> program
> in Swarm. I found CFS-C, via ftp, at lumpi.informatik.uni-dortmund.de; 
> 
>       at 
> 
> http://www.wior.uni-karlsruhe.de/Bibliothek/Software_for_OR/Genetic_Algorith
> ms/pub/CFS_C.html 
> 
> the program is quoted as "still the state-of-the-art implementation
> concerning CFS."
> 
>       Is anyone doing the same job? Is Rick planning to produce a new version
> or, better, to embed it in Swarm?
> 
>       Our work, if successfull, will be posted at Santa Fe site.
> 
>       Pietro
> 
> 
>                   ==================================
>    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.
>                   ==================================
> 

                  ==================================
   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.
                  ==================================


reply via email to

[Prev in Thread] Current Thread [Next in Thread]