clusters satisfy the locality property that the points in the same cluster are close to each other. A number of
clustering problems arise in machine learning where the optimal clusters do not follow such a locality property.
For instance, consider the r-gather clustering problem where there is an additional constraint that each of the
clusters should have at least r points or the capacitated clustering problem where there is an upper bound on
the cluster sizes. In this work, we consider an extreme generalisation of such problems by assuming that the
target clustering is an arbitrary partition of the dataset. We show some interesting results in this scenario.
This is joint work with Anup Bhattacharya (IIT Delhi) and Amit Kumar (IIT Delhi).
(arXiv link: http://arxiv.org/abs/1504.02564)