Novel Algorithm for Incremental L1-Norm Principal-Component Analysis

被引：0
|
作者：
Dhanaraj, Mayur [1 ]
Markopoulos, Panos P. [1 ]
机构：
[1] Rochester Inst Technol, Dept Elect & Microelect Engn, Rochester, NY 14623 USA
来源：
2018 26TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO) | 2018年
关键词：
Image/video processing; incremental PCA; L1-norm PCA; outliers; online learning; FOREGROUND EXTRACTION;
D O I：
暂无
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
L1-norm Principal-Component Analysis (L1-PCA) has been shown to exhibit sturdy resistance against outliers among the processed data. In this work, we propose L1-IPCA: an algorithm for incremental L1-PCA, appropriate for big-data and streaming-data applications. The proposed algorithm updates the calculated L1-norm principal components as new data points arrive, conducting a sequence of computationally efficient bit-flipping iterations. Our experimental studies on subspace estimation, image conditioning, and video foreground extraction illustrate that the proposed algorithm attains remarkable outlier resistance at low computational cost.
引用
收藏
页码：2020 / 2024
页数：5
相关论文
共 50 条
  • [1] Detection of Change by L1-norm Principal-Component Analysis
    Gallone, Ginevra
    Varma, Kavita
    Pados, Dimitris A.
    Colonnese, Stefania
    [J]. BIG DATA II: LEARNING, ANALYTICS, AND APPLICATIONS, 2020, 11395
  • [2] L1-Norm Principal-Component Analysis of Complex Data
    Tsagkarakis, Nicholas
    Markopoulos, Panos P.
    Sklivanitis, George
    Pados, Dimitris A.
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2018, 66 (12) : 3256 - 3267
  • [3] OPTIMAL SPARSE L1-NORM PRINCIPAL-COMPONENT ANALYSIS
    Chamadia, Shubham
    Pados, Dimitris A.
    [J]. 2017 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2017, : 2686 - 2690
  • [4] L1-norm Principal-Component Analysis via Bit Flipping
    Markopoulos, Panos P.
    Kundu, Sandipan
    Chamadia, Shubham
    Pados, Dimitris A.
    [J]. 2016 15TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA 2016), 2016, : 326 - 332
  • [5] Conformity evaluation and L1-norm principal-component analysis of tensor data
    Tountas, Konstantinos
    Pados, Dimitris A.
    Medley, Michael J.
    [J]. BIG DATA: LEARNING, ANALYTICS, AND APPLICATIONS, 2019, 10989
  • [6] Conformity Evaluation of Data Samples by L1-Norm Principal-Component Analysis
    Liu, Ying
    Pados, Dimitris A.
    [J]. COMPRESSIVE SENSING VII: FROM DIVERSE MODALITIES TO BIG DATA ANALYTICS, 2018, 10658
  • [7] Adaptive L1-Norm Principal-Component Analysis With Online Outlier Rejection
    Markopoulos, Panos P.
    Dhanaraj, Mayur
    Savakis, Andreas
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING, 2018, 12 (06) : 1131 - 1143
  • [8] Efficient L1-Norm Principal-Component Analysis via Bit Flipping
    Markopoulos, Panos P.
    Kundu, Sandipan
    Chamadia, Shubham
    Pados, Dimitris A.
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2017, 65 (16) : 4252 - 4264
  • [9] Iterative Re-weighted L1-Norm Principal-Component Analysis
    Liu, Ying
    Pados, Dimitris A.
    Batalama, Stella N.
    Medley, Michael J.
    [J]. 2017 FIFTY-FIRST ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS, AND COMPUTERS, 2017, : 425 - 429
  • [10] AN EFFICIENT ALGORITHM FOR L1-NORM PRINCIPAL COMPONENT ANALYSIS
    Yu, Linbin
    Zhang, Miao
    Ding, Chris
    [J]. 2012 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2012, : 1377 - 1380
12345