Comment Re:It's the fitness function, stupid... (Score 1) 174
yeah, i agree: this ga stuff is really cool but when you want to solve high level problems it gets really complicated (fitness function wise). so i dont expect to see a computer generated kernel anytime soon. but there are two problems where it looks like they could help: process schedulers and packet routing. those are NP complete problems,
arent they? and wouldn't it be cool to dynamically generate the schedulling algorithms in the kernel? you could let the user define a criteria and then let the computer evolve the algorithms. same would apply to routing. i havent STW on it, is this feasible? is anyone working on this?
arent they? and wouldn't it be cool to dynamically generate the schedulling algorithms in the kernel? you could let the user define a criteria and then let the computer evolve the algorithms. same would apply to routing. i havent STW on it, is this feasible? is anyone working on this?
