Simplified intelligence single particle optimization based neural network for digit recognition

Jiarui Zhou; Zhen Ji; Linlin Shen

doi:10.1109/CCPR.2008.74

Simplified intelligence single particle optimization based neural network for digit recognition

Jiarui Zhou, Zhen Ji, Linlin Shen

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

18 Citations (Scopus)

Abstract

To overcome the drawback of overly dependence on the input parameters in intelligence single particle optimization (ISPO), an improved algorithm, called simplified intelligence single particle optimization (SISPO), is proposed in this paper. While maintaining similar performance as ISPO, no special parameter settings are required by SISPO. The proposed SISPO was successfully applied to train neural network classifier for digit recognition. Experimental results demonstrated that, the proposed neural network training algorithm, Simplified Intelligence Single Particle Optimization Neural Network (SISPONN), achieved less training error and test error than traditional BP algorithms like gradient methods.

Original language	English
Title of host publication	Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008
Pages	349-353
Number of pages	5
DOIs	https://doi.org/10.1109/CCPR.2008.74
Publication status	Published - 2008
Externally published	Yes
Event	2008 Chinese Conference on Pattern Recognition, CCPR 2008 - Beijing, China Duration: 22 Oct 2008 → 24 Oct 2008

Publication series

Name	Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008

Conference

Conference	2008 Chinese Conference on Pattern Recognition, CCPR 2008
Country/Territory	China
City	Beijing
Period	22/10/08 → 24/10/08

Keywords

Digital recognition
ISPO
Intelligence single particle optimization
Neural network

ASJC Scopus subject areas

Computer Vision and Pattern Recognition

Access to Document

10.1109/CCPR.2008.74

Cite this

@inproceedings{81ec0c4e1a5b4d70982d703cdd67afa0,

title = "Simplified intelligence single particle optimization based neural network for digit recognition",

abstract = "To overcome the drawback of overly dependence on the input parameters in intelligence single particle optimization (ISPO), an improved algorithm, called simplified intelligence single particle optimization (SISPO), is proposed in this paper. While maintaining similar performance as ISPO, no special parameter settings are required by SISPO. The proposed SISPO was successfully applied to train neural network classifier for digit recognition. Experimental results demonstrated that, the proposed neural network training algorithm, Simplified Intelligence Single Particle Optimization Neural Network (SISPONN), achieved less training error and test error than traditional BP algorithms like gradient methods.",

keywords = "Digital recognition, ISPO, Intelligence single particle optimization, Neural network",

author = "Jiarui Zhou and Zhen Ji and Linlin Shen",

year = "2008",

doi = "10.1109/CCPR.2008.74",

language = "English",

isbn = "9781424423163",

series = "Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008",

pages = "349--353",

booktitle = "Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008",

note = "2008 Chinese Conference on Pattern Recognition, CCPR 2008 ; Conference date: 22-10-2008 Through 24-10-2008",

}

Zhou, J, Ji, Z & Shen, L 2008, Simplified intelligence single particle optimization based neural network for digit recognition. in Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008., 4663027, Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008, pp. 349-353, 2008 Chinese Conference on Pattern Recognition, CCPR 2008, Beijing, China, 22/10/08. https://doi.org/10.1109/CCPR.2008.74

Simplified intelligence single particle optimization based neural network for digit recognition. / Zhou, Jiarui; Ji, Zhen; Shen, Linlin.
Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008. 2008. p. 349-353 4663027 (Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Simplified intelligence single particle optimization based neural network for digit recognition

AU - Zhou, Jiarui

AU - Ji, Zhen

AU - Shen, Linlin

PY - 2008

Y1 - 2008

N2 - To overcome the drawback of overly dependence on the input parameters in intelligence single particle optimization (ISPO), an improved algorithm, called simplified intelligence single particle optimization (SISPO), is proposed in this paper. While maintaining similar performance as ISPO, no special parameter settings are required by SISPO. The proposed SISPO was successfully applied to train neural network classifier for digit recognition. Experimental results demonstrated that, the proposed neural network training algorithm, Simplified Intelligence Single Particle Optimization Neural Network (SISPONN), achieved less training error and test error than traditional BP algorithms like gradient methods.

AB - To overcome the drawback of overly dependence on the input parameters in intelligence single particle optimization (ISPO), an improved algorithm, called simplified intelligence single particle optimization (SISPO), is proposed in this paper. While maintaining similar performance as ISPO, no special parameter settings are required by SISPO. The proposed SISPO was successfully applied to train neural network classifier for digit recognition. Experimental results demonstrated that, the proposed neural network training algorithm, Simplified Intelligence Single Particle Optimization Neural Network (SISPONN), achieved less training error and test error than traditional BP algorithms like gradient methods.

KW - Digital recognition

KW - ISPO

KW - Intelligence single particle optimization

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=57949098192&partnerID=8YFLogxK

U2 - 10.1109/CCPR.2008.74

DO - 10.1109/CCPR.2008.74

M3 - Conference contribution

AN - SCOPUS:57949098192

SN - 9781424423163

T3 - Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008

SP - 349

EP - 353

BT - Proceedings of the 2008 Chinese Conference on Pattern Recognition, CCPR 2008

T2 - 2008 Chinese Conference on Pattern Recognition, CCPR 2008

Y2 - 22 October 2008 through 24 October 2008

ER -

Simplified intelligence single particle optimization based neural network for digit recognition

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this