Sparse nonnegative matrix approximation: new formulations and algorithms

2010

Technical Report

ei

We introduce several new formulations for sparse nonnegative matrix approximation. Subsequently, we solve these formulations by developing generic algorithms. Further, to help selecting a particular sparse formulation, we briefly discuss the interpretation of each formulation. Finally, preliminary experiments are presented to illustrate the behavior of our formulations and algorithms.

@techreport{6764,
  title = {Sparse nonnegative matrix approximation: new formulations and algorithms},
  author = {Tandon, R. and Sra, S.},
  number = {193},
  organization = {Max-Planck-Gesellschaft},
  institution = {Max Planck Institute for Biological Cybernetics, Tübingen, Germany},
  school = {Biologische Kybernetik},
  month = sep,
  year = {2010},
  doi = {},
  month_numeric = {9}
}