|
From: Nick C. <ntc...@ve...> - 2004-04-01 13:21:19
|
Jeff, Well, both approaches would be useful, I think. I had to pick I guess I'd go with the agent level one. If you post this query on repast-interest, you may get more responses as the number of subscribers is much larger. Nick > -----Original Message----- > From: rep...@li... > [mailto:rep...@li...]On Behalf Of > Jeffrey Fuller > Sent: Tuesday, March 30, 2004 10:07 PM > To: rep...@li... > Subject: [Repast-developer] Genetic Algorithm > > > In the process of coming up with a GA implementation for Repast, I have > come up one the question, what level make the most sense. Essentially, > I have devised three different levels that could all make sense in > their own way. At the lowest level, you could setup a GA that runs for > each agent evolving an ideal set of behavior values. The next level up > would evolve the population of agents in order to achieve a "perfect" > agent. The option would be to have a GA consist of a population of > models, which would yield a best model. > > What level you think would make the most sense for implementation into > Repast? > > > -Jeff > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: IBM Linux Tutorials > Free Linux tutorial presented by Daniel Robbins, President and CEO of > GenToo technologies. Learn everything from fundamentals to system > administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click > _______________________________________________ > Repast-developer mailing list > Rep...@li... > https://lists.sourceforge.net/lists/listinfo/repast-developer > |