English

Sparse Logistic Tensor Decomposition for Binary Data

Applications 2021-06-30 v1 Computation

Abstract

Tensor data are increasingly available in many application domains. We develop several tensor decomposition methods for binary tensor data. Different from classical tensor decompositions for continuous-valued data with squared error loss, we formulate logistic tensor decompositions for binary data with a Bernoulli likelihood. To enhance the interpretability of estimated factors and improve their stability further, we propose sparse formulations of logistic tensor decomposition by considering 1\ell_{1}-norm and 0\ell_{0}-norm regularized likelihood. To handle the resulting optimization problems, we develop computational algorithms which combine the strengths of tensor power method and majorization-minimization (MM) algorithm. Through simulation studies, we demonstrate the utility of our methods in analysis of binary tensor data. To illustrate the effectiveness of the proposed methods, we analyze a dataset concerning nations and their political relations and perform co-clustering of estimated factors to find associations between the nations and political relations.

Keywords

Cite

@article{arxiv.2106.14258,
  title  = {Sparse Logistic Tensor Decomposition for Binary Data},
  author = {Jianhao Zhang and Yoonkyung Lee},
  journal= {arXiv preprint arXiv:2106.14258},
  year   = {2021}
}