I have a multi-variate convex function, and I want to find its global minimum. We know there is one and only one minimum.
Is there any algorithm which is guaranteed to converge to the minimum starting from an arbitrary initial point?
How about the Newton method?
ps. The function is smooth.