A Block Coordinate Descent-Based Projected Gradient Algorithm for Orthogonal Non-Negative Matrix Factorization
This article uses the projected gradient method (PG) for a non-negative matrix factorization problem (NMF), where one or both matrix factors must have orthonormal columns or rows. We penalize the orthonormality constraints and apply the PG method via a block coordinate descent approach. This means t...
Main Authors: | Soodabeh Asadi, Janez Povh |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/9/5/540 |
Similar Items
-
Block Coordinate Descent for Regularized Multi-convex Optimization
by: Xu, Yangyang
Published: (2013) -
NOMA Codebook Optimization by Batch Gradient Descent
by: Zhongwei Si, et al.
Published: (2019-01-01) -
Efficient Asynchronous Semi-Stochastic Block Coordinate Descent Methods for Large-Scale SVD
by: Fanhua Shang, et al.
Published: (2021-01-01) -
Off-Grid DOA Estimation Using Alternating Block Coordinate Descent in Compressed Sensing
by: Weijian Si, et al.
Published: (2015-08-01) -
Aperture Shape Generation Based on Gradient Descent With Momentum
by: Liyuan Zhang, et al.
Published: (2019-01-01)