Genetic algorithms are passe, right? I mean, we've all moved on to MCMC because you get a more robust mathematical/statistical understanding of the underlying parameter space, and you can get out of local minima easily. I guess I never really understood what all the fuss was. If I know A is a better solution than B and C, then MCMC seems more reasonable: "jump from B to A most of the time, but occasionally jump to C, just to check that A isn't a local peak." Saying "make a child between A&B and A&C, and see which one is better" seems dumb, especially since you should know something about the gradients in most cases.
- This is kind of interesting, although people who do "art" like this are pretty much the extreme of pretentiousness.
- Chespin is best.
- Squirrel.
No comments:
Post a Comment