A Stochastic Nonconvex Splitting Method for Symmetric Nonnegative Matrix Factorization

Abstract

where (a) because that E and Tr are both linear operators and Tr[E[ZZ ]] is certain positive number which is independent on the optimization variables, then this term can be replaced by another positive constant Tr[EZ[Z ]EZ[Z]]. Based on the equivalence between (13) and (14), we can apply the result shown in [28, Lemma 2]. Since the constraint set is… (More)

Topics

3 Figures and Tables

Cite this paper

@inproceedings{Lu2017ASN, title={A Stochastic Nonconvex Splitting Method for Symmetric Nonnegative Matrix Factorization}, author={Songtao Lu and Mingyi Hong and Zhengdao Wang}, booktitle={AISTATS}, year={2017} }