An adaptive optimization method for estimating the number of components in a Gaussian mixture model

被引：1
|
作者：
Sun, Shuping [1 ]
Tong, Yaonan [1 ]
Zhang, Biqiang [2 ]
Yang, Bowen [3 ]
He, Peiguang [2 ]
Song, Wei [1 ]
Yang, Wenbo [2 ]
Wu, Yilin [4 ]
Liu, Guangyu [2 ]
机构：
[1] Hunan Inst Sci & Technol, Sch Informat Sci & Engn, Yueyang 414006, Peoples R China
[2] Nanyang Inst Technol, Dept Informat Engn, Nanyang 473004, Peoples R China
[3] Univ Chinese Acad Sci UCAS, Sch Integrated Circuits, Beijing 101400, Peoples R China
[4] Nanyang Inst Technol, Dept Intelligent Mfg, Nanyang 473004, Peoples R China
来源：
JOURNAL OF COMPUTATIONAL SCIENCE | 2022年 / 64卷
关键词：
GMM; MIGMM; chi(2) distribution; Mahalanobis distance; Adaptive optimal number; Adaptive interval; IMPROVED EM ALGORITHM; INFORMATION CRITERION; ORDER SELECTION; CREDIT RISK; K-MEANS; IDENTIFICATION; PREDICTION; APPROXIMATION; SYSTEMS;
D O I：
10.1016/j.jocs.2022.101874
中图分类号：
TP39 [计算机的应用];
学科分类号：
081203 ; 0835 ;
摘要：
Regarding the determination of the number of components (M) in a Gaussian mixture model (GMM), this study proposes a novel method for adaptively locating an optimal value of M when using a GMM to fit a given dataset; this method avoids underfitting and overfitting due to an unreasonable manually specified interval. The major contributions of this study are highlighted: (1) An adaptive interval for M (denoted as M is an element of [M-Min(Ada), M-Max(Ada)]) based on two procedures of a novel method, the modified incremental Gaussian mixture model (MIGMM), is determined via an adjustable parameter beta. (2) Considering some typical criteria, the optimal number.. within the obtained adaptive interval [M-Min(Ada), M-Max(Ada)], M-Opt(Ada) , is ultimately determined. Regarding the adaptive interval, extensive experiments with typical synthetic datasets show that [M-Min(Ada) M-Max(Ada)], corresponding to the parameter [beta(Min) = 10(-11), beta(Max) = 10(-2)], is determined. The performance of the M-Opt(Ada) determination based on several typical criteria is evaluated on both synthetic and real-world datasets.
引用
收藏
页数：15
相关论文
共 50 条
  • [21] A self-adaptive Gaussian mixture model
    Chen, Zezhi
    Ellis, Tim
    COMPUTER VISION AND IMAGE UNDERSTANDING, 2014, 122 : 35 - 46
  • [22] Adaptive method for segmentation of vehicles through local threshold in the Gaussian Mixture Model
    Buzar Lima, Kalyf Abdalla
    Teixeira Aires, Kelson Romulo
    Pereira dos Reis, Francisco Wender
    2015 BRAZILIAN CONFERENCE ON INTELLIGENT SYSTEMS (BRACIS 2015), 2015, : 204 - 209
  • [23] A GAUSSIAN MIXTURE MODEL-BASED REGULARIZATION METHOD IN ADAPTIVE IMAGE RESTORATION
    Liu Peng Zhang Yan Mao Zhigang (Shenzhen Graduate School
    Journal of Electronics(China), 2007, (01) : 83 - 89
  • [24] Image denoising method based on adaptive Gaussian mixture model in wavelet domain
    College of Communication Engineering, Jilin University, Changchun 130022, China
    Jilin Daxue Xuebao (Gongxueban), 2006, 6 (983-988):
  • [25] Optimization of the Gaussian Mixture Model Evaluation on GPU
    Vanek, Jan
    Trmal, Jan
    Psutka, Josef V.
    Psutka, Josef
    12TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2011 (INTERSPEECH 2011), VOLS 1-5, 2011, : 1748 - 1751
  • [26] Initializing the EM algorithm in Gaussian mixture models with an unknown number of components
    Melnykov, Volodymyr
    Melnykov, Igor
    COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2012, 56 (06) : 1381 - 1395
  • [27] A New Algorithm for Reducing Components of a Gaussian Mixture Model
    Yokoyama, Naoya
    Azuma, Daiki
    Tsukiyama, Shuji
    Fukui, Masahiro
    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 2016, E99A (12) : 2425 - 2434
  • [28] Estimation of Optimal Number of Components in Gaussian Mixture Model-Based Probabilistic Load Flow Study
    Prusty, B. Rajanarayan
    Jena, Debashisha
    2016 IEEE ANNUAL INDIA CONFERENCE (INDICON), 2016,
  • [29] ON ESTIMATING THE NUMBER OF COMPONENTS IN A FINITE MIXTURE OF POWER-SERIES DISTRIBUTIONS
    ANTONIADIS, A
    BERRUYER, J
    COMPUTATIONAL STATISTICS & DATA ANALYSIS, 1986, 4 (04) : 229 - 241
  • [30] Gaussian mixture model learning based image denoising method with adaptive regularization parameters
    Zhang, Jianwei
    Liu, Jing
    Li, Tong
    Zheng, Yuhui
    Wang, Jin
    MULTIMEDIA TOOLS AND APPLICATIONS, 2017, 76 (09) : 11471 - 11483
12345