Recursive unsupervised learning of finite mixture models

被引：170
|
作者：
Zivkovic, Z
van der Heijden, F
机构：
[1] Univ Amsterdam, Inst Informat, NL-1098 SJ Amsterdam, Netherlands
[2] Univ Twente, Lab Measurement & Instrumentat, NL-7500 AE Enschede, Netherlands
来源：
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 2004年 / 26卷 / 05期
关键词：
online (recursive) estimation; unsupervised learning; finite mixtures; model selection; EM-algorithm;
D O I：
10.1109/TPAMI.2004.1273970
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
There are two open problems when finite mixture densities are used to model multivariate data: the selection of the number of components and the initialization. In this paper, we propose an online (recursive) algorithm that estimates the parameters of the mixture and that simultaneously selects the number of components. The new algorithm starts with a large number of randomly initialized components. A prior is used as a bias for maximally structured models. A stochastic approximation recursive learning algorithm is proposed to search for the maximum a posteriori (MAP) solution and to discard the irrelevant components.
引用
收藏
页码：651 / 656
页数：6
相关论文
共 50 条
  • [1] Unsupervised learning of finite mixture models
    Figueiredo, MAT
    Jain, AK
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2002, 24 (03) : 381 - 396
  • [2] Unsupervised Greedy Learning of Finite Mixture Models
    Greggio, Nicola
    Bernardino, Alexandre
    Laschi, Cecilia
    Dario, Paolo
    Santos-Victor, Jose
    [J]. 22ND INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE (ICTAI 2010), PROCEEDINGS, VOL 2, 2010, : 219 - 224
  • [3] An algorithm for unsupervised learning and optimization of finite mixture models
    Abas, Ahmed R.
    [J]. EGYPTIAN INFORMATICS JOURNAL, 2011, 12 (01) : 19 - 27
  • [4] An unsupervised learning algorithm for image segmentation based on finite mixture models
    Yu, LS
    Zhang, TW
    [J]. 2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING, 2005, : 101 - 104
  • [5] Unsupervised learning for finite mixture models via modified gibbs sampling
    Liu, Weifeng
    Han, Chongzhao
    Shi, Yong
    [J]. Hsi-An Chiao Tung Ta Hsueh/Journal of Xi'an Jiaotong University, 2009, 43 (02): : 15 - 19
  • [6] Unsupervised selection and estimation of finite mixture models
    Figueiredo, MAT
    Jain, AK
    [J]. 15TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 2, PROCEEDINGS: PATTERN RECOGNITION AND NEURAL NETWORKS, 2000, : 87 - 90
  • [7] Anomaly Detection in IDSs by means of unsupervised greedy learning of finite mixture models
    Nicola Greggio
    [J]. Soft Computing, 2018, 22 : 3357 - 3372
  • [8] Anomaly Detection in IDSs by means of unsupervised greedy learning of finite mixture models
    Greggio, Nicola
    [J]. SOFT COMPUTING, 2018, 22 (10) : 3357 - 3372
  • [9] Unsupervised clustering using nonparametric finite mixture models
    Hunter, David R. R.
    [J]. WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL STATISTICS, 2024, 16 (01)
  • [10] Unsupervised Learning Using Variational Inference on Finite Inverted Dirichlet Mixture Models with Component Splitting
    Maanicshah, Kamal
    Amayri, Manar
    Bouguila, Nizar
    Fan, Wentao
    [J]. WIRELESS PERSONAL COMMUNICATIONS, 2021, 119 (02) : 1817 - 1844
12345