On a CSCI note, after listening to one of my professor's give a talk last semester about Genetic Programming (he did his Ph.D. on Genetic Programming where I believe he did symbolic regression to get a model for the GDP that was pretty darn accurate to what economics people use), he recommended a book Genetic Programming: On the Programming of Computers by Means of Natural Selection (Complex Adaptive Systems). I bought it a while back for a great price (had water damage so I ripped off the rotting cover and scotch taped it up, good as new) and I finally got around to reading through it while i've been off and I must say, that is some really cool stuff in there.
It still seems like black magic in a way, but I would love to either try something on my own or possibly do some research with my professor if he'd let me to get my feet wet in actually doing some Genetic Programming. The field still seems very new and the possibilities of its applications seem virtually unlimited.