Yes. Gradient descent with random restarts is the pragmatists first optimizations technique. GA's often don't converge to anything useful and require tons of un

Yes. Gradient descent with random restarts is the pragmatists first optimizations technique. GA's often don't converge to anything useful and require tons of unsubstantiated unintuiative tuning.