Algorithms, Initializations, and Convergence for the Nonnegative Matrix Factorization

ABSTRACT

It is well-known that good initializations can improve the speed and accuracy of the solutions of many nonnegative matrix factorization A = WH (NMF) algorithms. Many NMF algorithms are sensitive with respect to the initialization of W or H or both. This is especially true of algorithms of the alternating least squares (ALS) type, including the two new ALS algorithms that are presented in this paper. We compare the results of six initialization procedures (two standard and four new) on our ALS algorithms. Lastly, we discuss the practical issue of choosing an appropriate convergence criterion.

JOURNAL

NCSU Technical Report Math 81706, To be submitted.

CO-AUTHORS

Russell Albright, James Cox, David Duling, Amy Langville, and Carl Meyer

THE PDF FILE

NMFInitAlgConv.pdf

Return To Abstracts