English

Probabilistic Clustering Using Maximal Matrix Norm Couplings

Machine Learning 2018-10-12 v1 Information Theory math.IT Machine Learning

Abstract

In this paper, we present a local information theoretic approach to explicitly learn probabilistic clustering of a discrete random variable. Our formulation yields a convex maximization problem for which it is NP-hard to find the global optimum. In order to algorithmically solve this optimization problem, we propose two relaxations that are solved via gradient ascent and alternating maximization. Experiments on the MSR Sentence Completion Challenge, MovieLens 100K, and Reuters21578 datasets demonstrate that our approach is competitive with existing techniques and worthy of further investigation.

Keywords

Cite

@article{arxiv.1810.04738,
  title  = {Probabilistic Clustering Using Maximal Matrix Norm Couplings},
  author = {David Qiu and Anuran Makur and Lizhong Zheng},
  journal= {arXiv preprint arXiv:1810.04738},
  year   = {2018}
}

Comments

Presented at 56th Annual Allerton Conference on Communication, Control, and Computing, 2018

R2 v1 2026-06-23T04:35:28.287Z