Non‐monotone projection gradient method for non‐negative matrix factorization

Non‐monotone projection gradient method for non‐negative matrix factorization

0.00 Avg rating0 Votes
Article ID: iaor20122497
Volume: 51
Issue: 3
Start Page Number: 1163
End Page Number: 1171
Publication Date: Apr 2012
Journal: Computational Optimization and Applications
Authors: , ,
Keywords: matrices, statistics: inference
Abstract:

Since Non‐negative Matrix Factorization (NMF) was first proposed over a decade ago, it has attracted much attention, particularly when applied to numerous data analysis problems. Most of the existing algorithms for NMF are based on multiplicative iterative and alternating least squares algorithms. However, algorithms based on the optimization method are few, especially in the case where two variables are derived at the same time. In this paper, we propose a non‐monotone projection gradient method for NMF and establish the convergence results of our algorithm. Experimental results show that our algorithm converges to better solutions than popular multiplicative update‐based algorithms.

Reviews

Required fields are marked *. Your email address will not be published.