GECCO'08

The actual reason for my trip to Atlanta was attending GECCO - the Genetic and Evolutionary Computation Conference 2008. I have published some work there and thus presented that work in form of 2 posters and a paper presentation at the Graduate Student Workshop. One paper was on protein pattern classifier learning [Paperin, 2008a], the other - a conceptual investigation into using my computational model of Gavrilets' Holey Fitnes Landscapes for preventing premature convergence in evolutionary optimisation algorithms [Paperin, 2008b].

The conference was a brilliant one, everyone I met there liked it a lot.

When I started going to academic conferences a few years ago, my strategy was to see as many talks as possible. As a result, I suffered through several days of listening to presentations from 8am till 6pm and could not really remember a single one on the next day. I now realise that if I attend 10 talks in a day I may really remember only one, but if I attend only 5, I may remember 3 of them.

So this time my strategy was to take it easy. I have registered to work as a student volunteer in order entertain myself during the day before the conference. This turned out to be a good decision, as Atlanta appeared not do be a very nice place to stroll around (see this post). Another side effect was that I got part of my registration fee returned in exchange for the volunteer work. I helped setting up the equipment and packing the conference bags on Friday before the conference and went out for a few drinks with another student working there. He came from Brazil and had to sell his guitar in order to be able to come to GECCO. When I heard that I felt not very comfortable, as a few minutes ago I had criticised the student travel support I received from Monash University, which is not very good at all, but my new friend was a living example that when we say "not very good" we usually mean a comparison with European universities, and a lot of people have it more difficult still. I may have had trouble finding money for travelling, but I did not end up having to sell my stuff.

Anyway, this guy, whose name was Aranildo, was interesting. He earned money as a musician while doing research and had a lot of cool things to tell. Unfortunately I had to call it an early night and had to go to the hostel to try to get some sleep (I had a horrible jet lag) and to prepare for the next day's talk.

It did not work out all that well with the sleeping. I am normally not so much affected by jet lags, but a 13 hours time difference (mind the daylight saving time) is a lot. I fell asleep too early, got up in the middle of the night and could not sleep any more.

Nevertheless, the talk went well. I received some good comments, but rather of a general kind than about the specifics of the paper. This was really to be expected, as applying my holey fitness landscape model developed for a biological simulation within an optimisation context really is a little far fetched and the work was hardly more that an investigation of feasibility (yes, it is feasible, but more work will have to be done to see if this will bring any real improvement against standard techniques; if you are interested, have a look at the paper [Paperin, 2008b]).

There were better papers in the workshop than my, papers that presented solid results for a concrete class of optimisation problems, in fact the paper that received the best paper award at the workshop was better that some of the papers presented at the main conference.

After the workshop, which went all day, I went out for a pint of local beer with a few PhD students from Florida ("Sweet Water" - a quite enjoyable beer, much better than what you expect from American beers after tasting the stuff they usually export). My fellows left early and I did not fancy going to the hostel in order to avoid falling asleep too early again. I went to the outside smokers' lounge of the conference hotel and started chatting to the people there.

Most of the people in the smokers "lounge" came for a large trade convention/exhibition that took place in Atlanta at that time. Here I met some "real Americans", i.e. not the kind of people who do research at a university, but the sort of people you expect Americans to be from the movies. Those, who vote for George Bush, and who seriously and proudly believe that they live in the only truly free country in the world, that Europe is communist, and that they do good to people by going around and throwing bombs at them in the name of free trade. I wanted to avoid a political discussion at any cost, but there were some other students there (not from the conference), and it sort-of started anyway. With these people, many of the prejudices against the US one may have when living in western Europe appear to be true big time. This post is about GECCO, not American politics, and I will not go into detail here, but it suffices to say that they are so brain-washed by their own propaganda, that they chat about politics in the same way as a religious person may chat to someone who attempts the unfruitful task of convincing them that their god (whichever it may be) does not exist. They repeat well established set phrases and if you present an argument that demonstrated a flaw in their logic, they pause, get confused for half a second, but than their thoroughly rinsed brains resolve the argument by overriding the logic apparatus, and they conclusively and with great conviction repeat whatever they said before, ignoring that you just explained them how it contradicts to what they said two minutes earlier. I should know better. One should not try to convince a fool that his priest says wrong. Everyone is entitled to their beliefs, foolish or not, whether it is in the greatness of some supernatural being, or in the greatness of some state, or, in fact, in the greatness of the scientific method (my religion). Never mind. I was at GECCO.

The late chat in the smokers lounge helped me to get into the local time zone, I got to bed late and slept through the night. Next day I was working as a volunteer again. I was at the reception desk and there was extremely little to do. I spent most of the time chatting to the ladies who were in charge of running the conference. It was quite interesting to find out a few things from behind the scene. I should say, that they, and others who were involved, did a very nice job organising GECCO this year. After work I managed to attend a genetic programming tutorial by John Koza and then the day was over. In the evening I went out for drinks with a lot of students from the conference. We were 20 at least, it was good fun and very interesting. There were people from everywhere, from all sorts of backgrounds and research areas. I had quite a few beers that night.

The next few days were similar in nature. I took care not to attend any talks before 10am, as I wanted to be fully awake for the talks I do attend. This strategy made me miss a few interesting presentations, but considering how many excellent talks I did see, I think I took the right approach. There were a lot of famous people and I listened and talked to a quite a few of them. Sadly, when you do research in Australia, you only see such names in journals and it is quite nice to be able to associate a face and a voice with a name.

I used the poster session to hand out as many leaflets for LiveGraph as I could along with presenting my poster. It was nice to see some interest. As for the posters, I had a few interesting talks about those as well, to my surprise more about the holey fitness landscape one than about the protein localisation paper. I thought the latter was more "tangible". Anyway, although my work was presented only as 2 posters out of many, I think it was worth it.

Every night we went out. There were heaps of graduate students from everywhere (as I said) and quite a few of them wanted to get to know each other over a few beers. We went to a Jazz bar that plaid country music, to a sports bar, to a cafe/pub, and to some other places. It was great and I met quite a few people with whom I'd like to stay in touch.

I took the emails from several people and promised to send the email list to all of them. As I have been travelling since then I still have not done that (sorry guys), but I hope to be able to do it within the next few days, so please hang on there!))

After the conference, a lot of people who were staying in the student dorms (accommodation organised by the conference for students) moved to the hostel I was staying all of the time, so there was some going out next day as well. I did some sight-seeing in Atlanta with Aranildo and Brian, an Irish fellow, (photos coming soon guys) and we went out in the evening with other people from the conference we ran into when walking through the downtown.

Overall, I am very happy that I went to GECCO'08, although I only had a workshop paper and 2 posters for the main event. It was a great scientific and social experience, and I met some people with whom I'd like to stay in touch. I would really like to go to GECCO'09 next year, but it does not quite fit into my research plan, nor into my financial possibilities, so I am not yet sure whether I will be able to. Time will tell.

Below there are a few pictures from the conference. If you are one of the people who were at the conference, drop me an email if you want me to send you an archive with all pictures I have.

Having spent 2 days in Atlanta after the conference, I got myself a car (with a limited amount of adventures) and set of to Knoxville. But that is another story..

Images: