4 ms·
So these are the means in the L^r norms, with r in the reals rather than the integers?
by leif 17y ago
So these are the means in the L^r norms, with r in the reals rather than the integers?
- sesqu 17y agoThat's what I got out of it too (well, substituting "natural centrality measures" for "means"), except it bothers me that for n dimensions, it looks like he uses L₂ for the vector norm. I'd much prefer argmin 𝐱( sum 𝐬∈Sⁿ sum j≤n |𝐬ⱼ-𝐱ⱼ|^i )