A column-wise update algorithm for nonnegative matrix factorization in Bregman divergence with an orthogonal constraint

被引:0
|
作者
Keigo Kimura
Mineichi Kudo
Yuzuru Tanaka
机构
[1] Hokkaido University,Graduate School of Information Science and Technology
来源
Machine Learning | 2016年 / 103卷
关键词
Orthogonal nonnegative matrix factorization; Orthogonal Factorization; Bregman Divergence; Column-wise Update;
D O I
暂无
中图分类号
学科分类号
摘要
Recently orthogonal nonnegative matrix factorization (ONMF), imposing an orthogonal constraint into NMF, has been attracting a great deal of attention. ONMF is more appropriate than standard NMF for a clustering task because the constrained matrix can be considered as an indicator matrix. Several iterative ONMF algorithms have been proposed, but they suffer from slow convergence because of their matrix-wise updating. In this paper, therefore, a column-wise update algorithm is proposed for speeding up ONMF. To make the idea possible, we transform the matrix-based orthogonal constraint into a set of column-wise orthogonal constraints. The algorithm is stated first with the Frobenius norm and then with Bregman divergence, both for measuring the degree of approximation. Experiments on one artificial and six real-life datasets showed that the proposed algorithms converge faster than the other conventional ONMF algorithms, more than four times in the best cases, due to their smaller numbers of iterations.
引用
收藏
页码:285 / 306
页数:21
相关论文
共 50 条
  • [41] On the convergence of multiplicative update algorithms for nonnegative matrix factorization
    Lin, Chih-Jen
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 2007, 18 (06): : 1589 - 1596
  • [42] Window Memory Layout Scheme for Alternate Row-Wise/Column-Wise Matrix Access
    Guo, Lei
    Tang, Yuhua
    Dou, Yong
    Lei, Yuanwu
    Ma, Meng
    Zhou, Jie
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2013, E96D (12): : 2765 - 2775
  • [43] A constrained sparse algorithm for nonnegative matrix factorization
    Collage of Computer and Info. Eng., Hohai Univ., Nanjing
    211100, China
    Sichuan Daxue Xuebao (Gongcheng Kexue Ban), 1600, 2 (108-111):
  • [44] Majorization-Minimization for Sparse Nonnegative Matrix Factorization With the β-Divergence
    Marmin, Arthur
    Goulart, Jose Henrique de Morais
    Fevotte, Cedric
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2023, 71 : 1435 - 1447
  • [46] TOWARDS COMPLEX NONNEGATIVE MATRIX FACTORIZATION WITH THE BETA-DIVERGENCE
    Magron, Paul
    Virtanen, Tuomas
    2018 16TH INTERNATIONAL WORKSHOP ON ACOUSTIC SIGNAL ENHANCEMENT (IWAENC), 2018, : 156 - 160
  • [47] Algorithms for Nonnegative Matrix Factorization with the Kullback-Leibler Divergence
    Hien, Le Thi Khanh
    Gillis, Nicolas
    JOURNAL OF SCIENTIFIC COMPUTING, 2021, 87 (03)
  • [48] Nonnegative matrix factorization and I-divergence alternating minimization
    Finesso, Lorenzo
    Spreij, Peter
    LINEAR ALGEBRA AND ITS APPLICATIONS, 2006, 416 (2-3) : 270 - 287
  • [49] A HYBRID ITERATIVE ALGORITHM FOR NONNEGATIVE MATRIX FACTORIZATION
    Soltuz, Stefan M.
    Wang, Wenwu
    Jackson, Philip J. B.
    2009 IEEE/SP 15TH WORKSHOP ON STATISTICAL SIGNAL PROCESSING, VOLS 1 AND 2, 2009, : 409 - 412
  • [50] Incremental Nonnegative Matrix Factorization with Sparseness Constraint for Image Representation
    Sun, Jing
    Wang, Zhihui
    Li, Haojie
    Sun, Fuming
    ADVANCES IN MULTIMEDIA INFORMATION PROCESSING - PCM 2018, PT II, 2018, 11165 : 351 - 360