Ye' old general discussion board. Basically, for everything that isn't covered elsewhere. Come here to shoot the breeze, shoot your mouth off, or whatever suits your fancy. This forum is not for asking programming related questions.
Ignore that last sentance of mine...in the previous post
Intelligence and Evolution are two completely different things. Your meshing them together. Things are not near as complicated as they seem sometimes.
I think...I said that right at the get go...
I don't believe evolution is a complex process. Rather it's a simple, mostly random event, which occurs over millions of years. If it was pre-meditated, I think, like software, life would be a lot more advanced and with little error.
p.s-I wouldn't say evolution and intelligence are two different things...rather intelligence is a characteristic or result of evolution.
"Meshing" is the word. The conflating of personal experiences with how evolution and computers work simply muddles things. Statements like "your genes might change enough through random intervention" show this clearly. "Your genes" don't change. Genes reproduce or they don't. Every time we reproduce the genes change. Factors like the environment effect the ability those genes to reproduce.
But reproduction happening over a million years, or a million instruction cycles inside a CPU, are nothing like your personal experience -- try as you might to mesh and muddle.
@Hockey: you believe evolution "does" this and that? Listen, in science, we do experiments and observations, make conclusions, put them to the test, have our conclusions peer-reviewed to decrease the chance of error (always possible), thus we build hypotheses which get published to a wider community, which does additional tests on our data, or tries to duplicate our experiment with different data, etc., etc. and out of this process a theory is born.
A theory (in scientific, not layman's terms) is our explanation on how a part of the world around us works. There's no place for belief at that point, only fact. (Which is not to say that beliefs are not useful at the preceding stages of the scientific process). If you're not sure how something works - instead of "believing" how it works, go read it up, that's the benefit of science!
Anyway, what you're talking about is two already well researched topics - genetic algorithms and neural networks. The first model the way evolution works - a "goal" is set in a system, with "mutable" agents trying to reach it and an outside "fitness" function that selects which agents should "die" and which should further "mutate". Neural networks model the way brains work at the physical level - and are quite successfull at that. I remember seeing a demo of a NN with ridiculously few neurons - something like 200 I think - which was able to learn how to steer a car without leaving the road.