A Bias Trick for Centered Robust Principal Component Analysis (Student Abstract)

被引：0

作者：

He, Baokun ^{[1
]}

Wan, Guihong ^{[1
]}

Schweitzer, Haim ^{[1
]}

机构：

[1] Univ Texas Dallas, 800 W Campbell Rd, Richardson, TX 75083 USA

来源：

THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE | 2020年 / 34卷

关键词：

PCA;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Outlier based Robust Principal Component Analysis (RPCA) requires centering of the non-outliers. We show a "bias trick" that automatically centers these non-outliers. Using this bias trick we obtain the first RPCA algorithm that is optimal with respect to centering.

引用

页码：13807 / 13808

页数：2

共 6 条

[1] [Anonymous], 2012, NIPS, DOI DOI 10.1109/TIT.2011.2173156
[2] He B., 2019, ARXIV191108024
[3] Robust PCA and classification in biosciences
Hubert, M
Engelen, S
[J]. BIOINFORMATICS, 2004, 20 (11) : 1728 - 1736
[4] Coherence Pursuit: Fast, Simple, and Robust Principal Component Analysis
Rahmani, Mostafa
Atia, George K.
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2017, 65 (23) : 6260 - 6275
[5] Outlier-Robust PCA: The High-Dimensional Case
Xu, Huan
Caramanis, Constantine
Mannor, Shie
[J]. IEEE TRANSACTIONS ON INFORMATION THEORY, 2013, 59 (01) : 546 - 572
[6] Zhang HY, 2015, AAAI CONF ARTIF INTE, P3143

← 1 →