During the semester, Professor Wolffe and I submitted a developer’s report regarding my work– incorporating genetic algorithm optimization in multi-agent models using JGAP and RePast–to SwarmFest. After about a week or so waiting for evaluation, we received notification that we were accepted to present a poster at the conference, and they also requested a full paper. That should be pretty slick. I am even trying to get funding for my trip the conference from GVSU. Luckily, it is being held at the University of Michigan this year, so costs are minimum. Anyhow, after I finish my work for the semester, I will begin feverishly incorporating another stage in my applications, and refine my cluster metrics. Also, I have the task of writing a full paper. I have received some references from one of the coordinators, so I am going to try to get a hold of those materials as well. It should be a busy set of weeks. The conference falls from May 9 – 11, which does not give us a lot of time. But, I think it is well within my abilities.

I’ve posted my abstract below, but you can read it at more at the following link:

SwarmFest Abstracts

Development Report: Incorporating a Genetic Algorithm Framework into a Multi-agent Modeling System

The purpose of this project was to embed intelligence into varying levels of a Swarm-like simulation system. In particular, it entailed the incorporation of the JGAP genetic algorithm framework into the RePast agent-modeling tool.

In the course of integrating the framework into the tool, we identified three separate levels at which we could assimilate evolutionary computing-based intelligence. A genetic algorithm can run at the individual agent level, used to select from among possible agent behaviors. Evolutionary pressure can also manifest at the model level, using the notion of fitness to drive optimization of agents. Finally, a genetic algorithm can be employed to generate populations of simulations, used to select towards an optimal model.

Each of these implementation levels has advantages and drawbacks, and applications for which they are best suited. In this paper, we describe the details of our implementations, characterize appropriate uses, present preliminary results incorporating our system into an existing model, and indicate potential directions for future research.

Well, I had been meaning to post about this for a while, but never actually got around to it. As some of your know, and some of you do not, I have been doing an independent study throughout the semester on swarm intelligence. I elected to do an oral presentation for Student Scholarship Day (SSD). The presentations actually went pretty well. Since my advisor, Professor Wolffe was going to have his Architecture class attend the presentations of students he advise, so I had to give two presentations. Originally the idea was that we would all be scheduled for the 1 – 2 PM time slots, the same time as the architecture class. However, it seems that the afternoon session did not begin until 1:20, which means that two of use were scheduled for 1:20, and one for 1:40. Being the madman that I am, I agreed to give a 1 PM presentation for CS people, and then again at 1:20 for the general audience, since that is when I was originally scheduled.

The first presentation went pretty well, I was exactly within my time frame without the assistance of a mediator, and I did not trip over my tongue too much. However, the one thing that I did make sure to fix for the second presentation was explaining the actions, and motives of the agents. The room had quite a few people I would say that at least half of the classroom was filled. Everyone in this session was CS/IS. Unfortunately, I do not thing the concepts sunk in too well, because I only received questions from Professor Wolffe. Anyhow, I figured that since so many people showed up that no one would be there for the second presentation. I was wrong.

Moments after people poured out in order to see the next presentation in another room; other people came in to see my presentation. I believe these people were from quite a bit different walks of life the CS/IS. There were a few professors in the class, mostly life science people I believe. And, there were actually quite a few students in the room. Perhaps I was a required presentation? That would be sweet. Anyhow, the weird thing about that was that there were more females there to view my presentation than in the entire CS/IS program. Carl was there I think he was the only CS/IS person there, well aside from the guy who stuck around to see the presentation a second time. Also, I think these people had a better understanding on the concepts employed here, because I received many more probing questions regarding my simulation and methods used to derived data. That was very cool.

It is 1 AM in the morning, and I have nothing to do. This is really weird. After having spent an entire semester putting in 16 hour days, the fact that I actually have this weekend off is mind blowing. I am literally twiddling my thumbs. I?m so used to sitting in front of my terminal writing code, or papers that it just feels like I should be writing something, which is why I am now. Generally, I think that you will find that the posts that I post after this happen before this post. So, this is actually a little funky, but that way that I hope people read the site is to start with the oldest post that they have not read yet, then work their way up. But, in all likelihood that is not the way that it works. Anyhow, I think the Break in Clouds post is the only one that it actually seems weird to read out of order. The others should be ok. Anyhow, I need to start writing those posts that I am talking about; otherwise they will just be a dream.

Prompted by a question posed by landw, I have determined that George Washington was the man in terms of American Presidents. Now, I realized that I may not be 100% accurate in my statements, because the facts are what I roughly remember from reading history books. Anyhow, let me explain why Washington was ?the man.?

Washington started out his military career while we were still under the tyrannous British control. But, he used this time to fight against the French, and really are there many more noble things? Ah? the French who fight so poorly, and surrender so readily. Anyhow, he continued on got military education and what not. Several years later, with his own command he defeats those filthy Frenchman.

And, perhaps more importantly, he was pivotal in the in our liberation from British tyranny. Not only did he defeat the British while under the Articles of Confederation, he also speared headed the efforts prompting the Constitutional Convention. He is soon there after elected the first president of the United States.

Washington did his time, and handled things well. Proving that he was not only a great military mind, but a great political one as well. Perhaps this is where I agree with him the most. You see, Washington warned against a two party system like the one that we currently endure. In his original plan, multiple people would run under different banners, the man with the most votes would be president, and the man with the second most would be vice president. Makes sense to me, and balances thing out better. I know that people thing that the California recall election was a mad house, but I think this is the way things are SUPPOSED to work. People are supposed to have diverse, and powerful choices. Even if most of the people running are total crackpot, the people should have that choice. Unfortunately, we are currently stuck with the two party systems, which is sad, because the group are essentially the same, and produce very similar candidates. Thus, we are in essence a modern monarchy of sorts, controlled by the corporate world. But, that topic is worth an entire post of it?s own one day.