ITR-Score algorithm: An efficient Trace ratio criterion based algorithm for supervised dimensionality reduction

Mingbo Zhao; Zhao Zhang; Tommy W.S. Chow

doi:10.1109/IJCNN.2011.6033213

ITR-Score algorithm: An efficient Trace ratio criterion based algorithm for supervised dimensionality reduction

Mingbo Zhao
, Zhao Zhang
, Tommy W.S. Chow

Department of Electrical Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

8 Citations (Scopus)

Abstract

Dimensionality reduction has been a fundamental tool when dealing with high-dimensional dataset. And trace ration optimization has been widely used in dimensionality reduction because Trace ratio can directly reflect the similarity (Euclidean distance) of data points. Conventionally, there is no close-form solution to the original trace ratio problem. Prior works have indicated that trace ratio problem can be solved by an iterative way. In this paper, we propose an efficient algorithm to find the optimal solutions. The proposed algorithm can be easily extended to its corresponding kernel version for handling the nonlinear problems. Finally, we evaluate our proposed algorithm based on extensive simulations of real world datasets. The results show our proposed method is able to deliver marked improvements over other supervised and unsupervised algorithms. © 2011 IEEE.

Original language	English
Title of host publication	Proceedings of the International Joint Conference on Neural Networks
Pages	145-152
DOIs	https://doi.org/10.1109/IJCNN.2011.6033213
Publication status	Published - 2011
Event	2011 International Joint Conference on Neural Network, IJCNN 2011 - San Jose, United States Duration: 31 Jul 2011 → 5 Aug 2011 https://neural.memberclicks.net/assets/docs/2011%20ijcnn%20program%20book.pdf

Conference

Conference	2011 International Joint Conference on Neural Network, IJCNN 2011
Place	United States
City	San Jose
Period	31/07/11 → 5/08/11
Internet address	https://neural.memberclicks.net/assets/docs/2011%20ijcnn%20program%20book.pdf

Research Keywords

Dimensionality reduction
Discriminative learning
Trace ratio criterion

Access Output

10.1109/IJCNN.2011.6033213

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{ed89bf45380246b5a846c5b17c416c85,

title = "ITR-Score algorithm: An efficient Trace ratio criterion based algorithm for supervised dimensionality reduction",

abstract = "Dimensionality reduction has been a fundamental tool when dealing with high-dimensional dataset. And trace ration optimization has been widely used in dimensionality reduction because Trace ratio can directly reflect the similarity (Euclidean distance) of data points. Conventionally, there is no close-form solution to the original trace ratio problem. Prior works have indicated that trace ratio problem can be solved by an iterative way. In this paper, we propose an efficient algorithm to find the optimal solutions. The proposed algorithm can be easily extended to its corresponding kernel version for handling the nonlinear problems. Finally, we evaluate our proposed algorithm based on extensive simulations of real world datasets. The results show our proposed method is able to deliver marked improvements over other supervised and unsupervised algorithms. {\textcopyright} 2011 IEEE.",

keywords = "Dimensionality reduction, Discriminative learning, Trace ratio criterion",

author = "Mingbo Zhao and Zhao Zhang and Chow, \{Tommy W.S.\}",

year = "2011",

doi = "10.1109/IJCNN.2011.6033213",

language = "English",

isbn = "9781457710865",

pages = "145--152",

booktitle = "Proceedings of the International Joint Conference on Neural Networks",

note = "2011 International Joint Conference on Neural Network, IJCNN 2011 ; Conference date: 31-07-2011 Through 05-08-2011",

url = "https://neural.memberclicks.net/assets/docs/2011\%20ijcnn\%20program\%20book.pdf",

}

Zhao, M, Zhang, Z & Chow, TWS 2011, ITR-Score algorithm: An efficient Trace ratio criterion based algorithm for supervised dimensionality reduction. in Proceedings of the International Joint Conference on Neural Networks., 6033213, pp. 145-152, 2011 International Joint Conference on Neural Network, IJCNN 2011, San Jose, California, United States, 31/07/11. https://doi.org/10.1109/IJCNN.2011.6033213

ITR-Score algorithm: An efficient Trace ratio criterion based algorithm for supervised dimensionality reduction. / Zhao, Mingbo; Zhang, Zhao; Chow, Tommy W.S.
Proceedings of the International Joint Conference on Neural Networks. 2011. p. 145-152 6033213.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - ITR-Score algorithm

T2 - 2011 International Joint Conference on Neural Network, IJCNN 2011

AU - Zhao, Mingbo

AU - Zhang, Zhao

AU - Chow, Tommy W.S.

PY - 2011

Y1 - 2011

N2 - Dimensionality reduction has been a fundamental tool when dealing with high-dimensional dataset. And trace ration optimization has been widely used in dimensionality reduction because Trace ratio can directly reflect the similarity (Euclidean distance) of data points. Conventionally, there is no close-form solution to the original trace ratio problem. Prior works have indicated that trace ratio problem can be solved by an iterative way. In this paper, we propose an efficient algorithm to find the optimal solutions. The proposed algorithm can be easily extended to its corresponding kernel version for handling the nonlinear problems. Finally, we evaluate our proposed algorithm based on extensive simulations of real world datasets. The results show our proposed method is able to deliver marked improvements over other supervised and unsupervised algorithms. © 2011 IEEE.

AB - Dimensionality reduction has been a fundamental tool when dealing with high-dimensional dataset. And trace ration optimization has been widely used in dimensionality reduction because Trace ratio can directly reflect the similarity (Euclidean distance) of data points. Conventionally, there is no close-form solution to the original trace ratio problem. Prior works have indicated that trace ratio problem can be solved by an iterative way. In this paper, we propose an efficient algorithm to find the optimal solutions. The proposed algorithm can be easily extended to its corresponding kernel version for handling the nonlinear problems. Finally, we evaluate our proposed algorithm based on extensive simulations of real world datasets. The results show our proposed method is able to deliver marked improvements over other supervised and unsupervised algorithms. © 2011 IEEE.

KW - Dimensionality reduction

KW - Discriminative learning

KW - Trace ratio criterion

UR - https://www.scopus.com/pages/publications/80054748404

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-80054748404&origin=recordpage

U2 - 10.1109/IJCNN.2011.6033213

DO - 10.1109/IJCNN.2011.6033213

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781457710865

SP - 145

EP - 152

BT - Proceedings of the International Joint Conference on Neural Networks

Y2 - 31 July 2011 through 5 August 2011

ER -

ITR-Score algorithm: An efficient Trace ratio criterion based algorithm for supervised dimensionality reduction

Abstract

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this