Juan J. Merelo wrote: > It's usually bounded, right? Not theoretically, but most experiments > limit tree depth.
Yes, sure. Unbounded Growth, Bad. :-) > nowadays, size-changing genomes are used (for instance, > evolving neural nets). Sure. > Which one would you vote for? Evolutionary computation is not purely an > artificial intelligence technique, that is why it's better to just leave > it in the Algorithm namespace. I would like to see a new top-level catetory created, EC::. AI encompasses too much stuff, and what it includes is extremely nebulous. In contrast, it's pretty easy to tell whether something can be classified as evolutionary, yet the field is still huge. Now keep in mind that if you have a framework (which I infer that you do), you may need your own special space for modules which can't reasonably be used outside the framework. In other words, don't monopolize EC:: with 20 modules that all have to work in concert... because I might want to drop a module or two in EC:: and there should be no confusion. -- John Douglas Porter I no longer love the color of your sweaters.
