Volume regularized non-negative matrix factorizations

Andersen Ang, M. S. and Gillis, Nicolas (2018) Volume regularized non-negative matrix factorizations. In 2018 9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2018. vol. 2018-September, IEEE Computer Society.. (doi:10.1109/WHISPERS.2018.8747250).

Record type: Conference or Workshop Item (Paper)

Abstract

This work considers two volume regularized non-negative matrix factorization (NMF) problems that decompose a nonnegative matrix X into the product of two nonnegative matrices W and H with a regularization on the volume of the convex hull spanned by the columns of W. This regularizer takes two forms: the determinant (det) and logarithm of the determinant (logdet) of the Gramian of W. In this paper, we explore the structure of these problems and present several algorithms, including a new algorithm based on an eigenvalue upper bound of the logdet function. Experimental results on synthetic data show that (i) the new algorithm is competitive with the standard Taylor bound, and (ii) the logdet regularizer works better than the det regularizer. We also illustrate the applicability of the new algorithm on the San Diego airport hyperspectral image.

This record has no associated files available for download.

More information

Published date: 23 September 2018

Venue - Dates: 9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2018, , Amsterdam, Netherlands, 2018-09-23 - 2018-09-26

Keywords: coordinate descent, determinant, log-determinant, Non-negative matrix factorization, volume regularizer

Learn more about Vision, Learning and Control research