Solving non-negative matrix factorization by alternating least squares with a modified strategy

被引：21
|
作者：
Liu, Hongwei [1 ]
Li, Xiangli [1 ,2 ]
Zheng, Xiuyun [1 ]
机构：
[1] Xidian Univ, Dept Math, Xian 710071, Peoples R China
[2] Guilin Univ Elect Technol, Coll Math & Comp Sci, Guilin 541004, Peoples R China
来源：
DATA MINING AND KNOWLEDGE DISCOVERY | 2013年 / 26卷 / 03期
基金：
中国国家自然科学基金;
关键词：
Non-negative matrix factorization; Alternating non-negative least squares; ALGORITHMS; DISCOVERY;
D O I：
10.1007/s10618-012-0265-y
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Non-negative matrix factorization (NMF) is a method to obtain a representation of data using non-negativity constraints. A popular approach is alternating non-negative least squares (ANLS). As is well known, if the sequence generated by ANLS has at least one limit point, then the limit point is a stationary point of NMF. However, no evdience has shown that the sequence generated by ANLS has at least one limit point. In order to overcome this shortcoming, we propose a modified strategy for ANLS in this paper. The modified strategy can ensure the sequence generated by ANLS has at least one limit point, and this limit point is a stationary point of NMF. The results of numerical experiments are reported to show the effectiveness of the proposed algorithm.
引用
收藏
页码：435 / 451
页数：17
相关论文
共 50 条
  • [1] Solving non-negative matrix factorization by alternating least squares with a modified strategy
    Hongwei Liu
    Xiangli Li
    Xiuyun Zheng
    Data Mining and Knowledge Discovery, 2013, 26 : 435 - 451
  • [2] Regularized alternating least squares algorithms for non-negative matrix/tensor factorization
    Cichocki, Andrzej
    Zdunek, Rafal
    ADVANCES IN NEURAL NETWORKS - ISNN 2007, PT 3, PROCEEDINGS, 2007, 4493 : 793 - +
  • [3] Hierarchical alternating nonlinear least squares for non-negative matrix factorization using rational functions
    Hautecoeur, Cecile
    Glineur, Francois
    De Lathauwer, Lieven
    29TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO 2021), 2021, : 1045 - 1049
  • [4] Research on the fast algorithm for hierarchic least squares of non-negative matrix factorization
    Xin, Huang
    Journal of Convergence Information Technology, 2012, 7 (11) : 359 - 368
  • [5] Cancer class discovery using non-negative matrix factorization based on alternating non-negativity-constrained least squares
    Kim, Hyunsoo
    Park, Haesun
    BIOINFORMATICS RESEARCH AND APPLICATIONS, PROCEEDINGS, 2007, 4463 : 477 - +
  • [6] A unified framework for sparse non-negative least squares using multiplicative updates and the non-negative matrix factorization problem
    Fedorov, Igor
    Nalci, Alican
    Giri, Ritwik
    Rao, Bhaskar D.
    Nguyen, Truong Q.
    Garudadri, Harinath
    SIGNAL PROCESSING, 2018, 146 : 79 - 91
  • [7] Monitor Ionizing Radiation-Induced Cellular Responses with Raman Spectroscopy, Non-Negative Matrix Factorization, and Non-Negative Least Squares
    Deng, Xinchen
    Ali-Adeeb, Ramie
    Andrews, Jeffrey L.
    Shreeves, Phillip
    Lum, Julian J.
    Brolo, Alexandre
    Jirasek, Andrew
    APPLIED SPECTROSCOPY, 2020, 74 (06) : 701 - 711
  • [8] Non-negative tensor factorization based on alternating large-scale non-negativity-constrained least squares
    Kim, Hyunsoo
    Park, Haesun
    Elden, Lars
    PROCEEDINGS OF THE 7TH IEEE INTERNATIONAL SYMPOSIUM ON BIOINFORMATICS AND BIOENGINEERING, VOLS I AND II, 2007, : 1147 - +
  • [9] Multichannel non-negative matrix factorisation based on alternating least squares for audio source separation system
    Lee, Seokjin
    Pang, Hee-Suk
    ELECTRONICS LETTERS, 2015, 51 (03) : 197 - 198
  • [10] Alternating Direction Method of Multipliers for Convolutive Non-Negative Matrix Factorization
    Li, Yinan
    Wang, Ruili
    Fang, Yuqiang
    Sun, Meng
    Luo, Zhangkai
    IEEE TRANSACTIONS ON CYBERNETICS, 2023, 53 (12) : 7735 - 7748
12345