The clustering algorithm splits points into disjoint sets in a way that minimizes the intra-set distance.
Is there an efficient algorithm which does the opposite (maximize the intra-set distance)? I.E. I would like the points in each set to be as "spread out" as possible.
I don’t get the question. To me that doesn’t make much sense. By definition if d is a distance and x a vector $d(x,x)=0$. How would you deal with it then ? x could not even belong to its own cluster.