Efficient source adaptivity in independent component analysis

被引：48
作者：
Vlassis, N
Motomura, Y
机构：
[1] Univ Amsterdam, RWCP, Autonomous Learning Funct SNN, Dept Comp Sci, NL-1098 SJ Amsterdam, Netherlands
[2] Electrotech Lab, Tsukuba, Ibaraki 3058568, Japan
来源：
IEEE TRANSACTIONS ON NEURAL NETWORKS | 2001年 / 12卷 / 03期
关键词：
blind signal separation; independent component analysis (ICA); score function estimation; source adaptivity;
D O I：
10.1109/72.925558
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
A basic element in most independent component analysis (ICA) algorithms is the choice of a model for the score functions of the unknown sources. While this is usually based on approximations, for large data sets it is possible to achieve "source adaptivity" by directly estimating from the data the "'true" score functions of the sources. In this paper we describe an efficient scheme for achieving this by extending the fast density estimation method of Silverman (1982), We show with a real and a synthetic experiment that our method can provide more accurate solutions than state-of-the-art methods when optimization is carried out in the vicinity of the global minimum of the contrast function.
引用
收藏
页码：559 / 566
页数：8
相关论文
共 50 条
[41]   Digital Image Stabilization by Independent Component Analysis [J].
Amanatiadis, Angelos A. ;
Andreadis, Ioannis .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2010, 59 (07) :1755-1763
[42]   Sequential Independent Component Analysis Density Estimation [J].
Aladjem, Mayer ;
Israeli-Ran, Itamar ;
Bortman, Maria .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (10) :5084-5097
[43]   Normalized natural gradient in independent component analysis [J].
Tang, Ying ;
Li, Jianping .
SIGNAL PROCESSING, 2010, 90 (09) :2773-2777
[44]   Ensemble learning approach to Independent Component Analysis [J].
Choudrey, R. ;
Penny, W.D. ;
Roberts, S.J. .
Neural Networks for Signal Processing - Proceedings of the IEEE Workshop, 2000, 1 :435-444
[45]   Vibration sources identification with independent component analysis [J].
Ye, Hongxian ;
Yang, Shixi ;
Yang, Jiangxin ;
Ji, Huawei .
WCICA 2006: SIXTH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-12, CONFERENCE PROCEEDINGS, 2006, :5814-+
[46]   CONSTRAINED INDEPENDENT COMPONENT ANALYSIS FOR HYPERSPECTRAL UNMIXING [J].
Xia, Wei ;
Wang, Bin ;
Zhang, Liming .
2010 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2010, :1293-1296
[47]   Independent component analysis using pots models [J].
Wu, JM ;
Chiu, SJ .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (02) :202-211
[48]   A "nonnegative PCA" algorithm for independent component analysis [J].
Plumbley, MD ;
Oja, E .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2004, 15 (01) :66-76
[49]   An EM algorithm for convolutive independent component analysis [J].
Deligne, S ;
Gopinath, R .
NEUROCOMPUTING, 2002, 49 :187-211
[50]   A new tool for the identification and localization of electromagnetic sources by using independent component analysis [J].
Capizzi, Giacomo ;
Coco, Salvatore ;
Laudani, Antonino .
IEEE TRANSACTIONS ON MAGNETICS, 2007, 43 (04) :1625-1628
12345