Quote:
Originally Posted by smokemon
[You must be logged in to view images. Log in or Register.]
|
I only read the abstract, and I already know that the paper does not say what you think it says. What your paper claims is that genetic algorithms are not better than other optimization algorithms like, say, simulated annealing. Now I am already suspicious of this - if you think about, this kind of thing is very difficult to prove empirically - but I can't be bothered to read the paper deeply enough to look at their methodology.
But it doesn't matter because you totally missed the point. ANY optimization algorithm is GREAT at finding local maxima and minima. The problem is whether you are at a global max/min. Species make perfect sense if you think of each one as a local maximum.