Robust tensor analysis with non-greedy ℓ1-norm maximization

Limei Zhao; Weimin Jia; Rong Wang; Qiang Yu

doi:10.13164/re.2016.0200

Robust tensor analysis with non-greedy ℓ₁-norm maximization

Limei Zhao, Weimin Jia, Rong Wang, Qiang Yu

Xi'an Research Institute of High Technology

Research output: Contribution to journal › Article › peer-review

12 Scopus citations

Abstract

The ℓ₁-norm based tensor analysis (TPCA-L1) is recently proposed for dimensionality reduction and feature extraction. However, a greedy strategy was utilized for solving the ℓ₁-norm maximization problem, which makes it prone to being stuck in local solutions. In this paper, we propose a robust TPCA with non-greedy ℓ₁-norm maximization (TPCA-L1 non-greedy), in which all projection directions are optimized simultaneously. Experiments on several face databases demonstrate the effectiveness of the proposed method.

Original language	English
Pages (from-to)	200-207
Number of pages	8
Journal	Radioengineering
Volume	25
Issue number	1
DOIs	https://doi.org/10.13164/re.2016.0200
State	Published - 1 Apr 2016
Externally published	Yes

Keywords

Non-greedy strategy
Outliers
Principal component analysis (PCA)
TPCA
ℓ-norm

Access to Document

10.13164/re.2016.0200

Cite this

@article{6c13d694cadb4e5a8102f8b9d34d9b20,

title = "Robust tensor analysis with non-greedy ℓ1-norm maximization",

abstract = "The ℓ1-norm based tensor analysis (TPCA-L1) is recently proposed for dimensionality reduction and feature extraction. However, a greedy strategy was utilized for solving the ℓ1-norm maximization problem, which makes it prone to being stuck in local solutions. In this paper, we propose a robust TPCA with non-greedy ℓ1-norm maximization (TPCA-L1 non-greedy), in which all projection directions are optimized simultaneously. Experiments on several face databases demonstrate the effectiveness of the proposed method.",

keywords = "Non-greedy strategy, Outliers, Principal component analysis (PCA), TPCA, ℓ-norm",

author = "Limei Zhao and Weimin Jia and Rong Wang and Qiang Yu",

year = "2016",

month = apr,

day = "1",

doi = "10.13164/re.2016.0200",

language = "英语",

volume = "25",

pages = "200--207",

journal = "Radioengineering",

issn = "1210-2512",

publisher = "Czech Technical University",

number = "1",

}

TY - JOUR

T1 - Robust tensor analysis with non-greedy ℓ1-norm maximization

AU - Zhao, Limei

AU - Jia, Weimin

AU - Wang, Rong

AU - Yu, Qiang

PY - 2016/4/1

Y1 - 2016/4/1

N2 - The ℓ1-norm based tensor analysis (TPCA-L1) is recently proposed for dimensionality reduction and feature extraction. However, a greedy strategy was utilized for solving the ℓ1-norm maximization problem, which makes it prone to being stuck in local solutions. In this paper, we propose a robust TPCA with non-greedy ℓ1-norm maximization (TPCA-L1 non-greedy), in which all projection directions are optimized simultaneously. Experiments on several face databases demonstrate the effectiveness of the proposed method.

AB - The ℓ1-norm based tensor analysis (TPCA-L1) is recently proposed for dimensionality reduction and feature extraction. However, a greedy strategy was utilized for solving the ℓ1-norm maximization problem, which makes it prone to being stuck in local solutions. In this paper, we propose a robust TPCA with non-greedy ℓ1-norm maximization (TPCA-L1 non-greedy), in which all projection directions are optimized simultaneously. Experiments on several face databases demonstrate the effectiveness of the proposed method.

KW - Non-greedy strategy

KW - Outliers

KW - Principal component analysis (PCA)

KW - TPCA

KW - ℓ-norm

UR - http://www.scopus.com/inward/record.url?scp=84964589418&partnerID=8YFLogxK

U2 - 10.13164/re.2016.0200

DO - 10.13164/re.2016.0200

M3 - 文章

AN - SCOPUS:84964589418

SN - 1210-2512

VL - 25

SP - 200

EP - 207

JO - Radioengineering

JF - Radioengineering

IS - 1

ER -

Robust tensor analysis with non-greedy ℓ₁-norm maximization

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this