Parallel Nonnegative Matrix Factorization via Newton Iteration-Reference-Cited by-同舟云学术

Parallel Nonnegative Matrix Factorization via Newton Iteration

Published:2016-09 Issue:03 Volume:26 Page:1650014
ISSN:0129-6264
Container-title:Parallel Processing Letters
language:en
Short-container-title:Parallel Process. Lett.

Author:

Flatz Markus¹,Vajteršic Marián¹²

Affiliation:

1. Department of Computer Sciences, University of Salzburg, Jakob-Haringer-Str. 2, 5020 Salzburg, Austria

2. Mathematical Institute, Department of Informatics, Slovak Academy of Sciences, Dúbravská cesta 9, 841 04 Bratislava, Slovakia

Abstract

The goal of Nonnegative Matrix Factorization (NMF) is to represent a large nonnegative matrix in an approximate way as a product of two significantly smaller nonnegative matrices. This paper shows in detail how an NMF algorithm based on Newton iteration can be derived using the general Karush-Kuhn-Tucker (KKT) conditions for first-order optimality. This algorithm is suited for parallel execution on systems with shared memory and also with message passing. Both versions were implemented and tested, delivering satisfactory speedup results.

Publisher

World Scientific Pub Co Pte Lt

Subject

Hardware and Architecture,Theoretical Computer Science,Software

Link

https://www.worldscientific.com/doi/pdf/10.1142/S0129626416500146

Reference9 articles.

1. Algorithms and applications for approximate nonnegative matrix factorization

2. Metagenes and molecular pattern discovery using matrix factorization

3. Improving molecular cancer class discovery through sparse non-negative matrix factorization

4. Super linear speedup in a local parallel meshless solution of thermo-fluid problems

Cited by 5 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Two-dimensional data partitioning for non-negative matrix tri-factorization;Big Data Research;2024-08

2. NMTF-LTM: Towards an Alignment of Semantics for Lifelong Topic Modeling;IEEE Transactions on Knowledge and Data Engineering;2023-10-01

3. Parallel Non-negative Matrix Tri-Factorization for Text Data Co-clustering;IEEE Transactions on Knowledge and Data Engineering;2022

4. Parallel Nonnegative Matrix Factorization Based on Newton Iteration with Improved Convergence Behavior;Parallel Processing and Applied Mathematics;2018

5. Improvements in Approximation Performance and Parallelization of Nonnegative Matrix Factorization with Newton Iteration;2017 International Conference on High Performance Computing & Simulation (HPCS);2017-07