A Unified Framework for Stochastic Matrix Factorization via Variance Reduction

05/19/2017
by   Renbo Zhao, et al.
0

We propose a unified framework to speed up the existing stochastic matrix factorization (SMF) algorithms via variance reduction. Our framework is general and it subsumes several well-known SMF formulations in the literature. We perform a non-asymptotic convergence analysis of our framework and derive computational and sample complexities for our algorithm to converge to an ϵ-stationary point in expectation. In addition, extensive experiments for a wide class of SMF formulations demonstrate that our framework consistently yields faster convergence and a more accurate output dictionary vis-à-vis state-of-the-art frameworks.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset