English

Scalable methods for nonnegative matrix factorizations of near-separable tall-and-skinny matrices

Machine Learning 2018-01-08 v1 Distributed, Parallel, and Cluster Computing Numerical Analysis Machine Learning

Abstract

Numerous algorithms are used for nonnegative matrix factorization under the assumption that the matrix is nearly separable. In this paper, we show how to make these algorithms efficient for data matrices that have many more rows than columns, so-called "tall-and-skinny matrices". One key component to these improved methods is an orthogonal matrix transformation that preserves the separability of the NMF problem. Our final methods need a single pass over the data matrix and are suitable for streaming, multi-core, and MapReduce architectures. We demonstrate the efficacy of these algorithms on terabyte-sized synthetic matrices and real-world matrices from scientific computing and bioinformatics.

Keywords

Cite

@article{arxiv.1402.6964,
  title  = {Scalable methods for nonnegative matrix factorizations of near-separable tall-and-skinny matrices},
  author = {Austin R. Benson and Jason D. Lee and Bartek Rajwa and David F. Gleich},
  journal= {arXiv preprint arXiv:1402.6964},
  year   = {2018}
}