Timezone: »

Near-Optimal Comparison Based Clustering
Michaël Perrot · Pascal Esser · Debarghya Ghoshdastidar

Tue Dec 08 09:00 AM -- 11:00 AM (PST) @ Poster Session 1 #509

The goal of clustering is to group similar objects into meaningful partitions. This process is well understood when an explicit similarity measure between the objects is given. However, far less is known when this information is not readily available and, instead, one only observes ordinal comparisons such as ``object i is more similar to j than to k.'' In this paper, we tackle this problem using a two-step procedure: we estimate a pairwise similarity matrix from the comparisons before using a clustering method based on semi-definite programming (SDP). We theoretically show that our approach can exactly recover a planted clustering using a near-optimal number of passive comparisons. We empirically validate our theoretical findings and demonstrate the good behaviour of our method on real data.

Author Information

Michaël Perrot (INRIA)
Pascal Esser (Technical University of Munich)
Debarghya Ghoshdastidar (Technical University Munich)

More from the Same Authors