Integration of magnified gradient function and weight evolution with deterministic perturbation into back-propagation

Sin-Chun Ng; Chi-Chung Cheung; Shu-Hung Leung

doi:10.1049/el:20030277

Integration of magnified gradient function and weight evolution with deterministic perturbation into back-propagation

Sin-Chun Ng, Chi-Chung Cheung, Shu-Hung Leung

Department of Electrical Engineering

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

1 Citation (Scopus)

Abstract

An approach for the improved performance of back-propagation (BP) learning systems, based on the integration of magnified gradient function and weight evolution with deterministic perturbation, was discussed. In this regard, the regression and character recognition problems were considered. The simulation results, in terms of the convergence rate and global convergence, showed that the integrated approach always outperformed other traditional methods.

Original language	English
Pages (from-to)	447-448
Journal	Electronics Letters
Volume	39
Issue number	5
DOIs	https://doi.org/10.1049/el:20030277
Publication status	Published - 6 Mar 2003

Access Output

10.1049/el:20030277

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{537d50821f074336b648917d7b996943,

title = "Integration of magnified gradient function and weight evolution with deterministic perturbation into back-propagation",

abstract = "An approach for the improved performance of back-propagation (BP) learning systems, based on the integration of magnified gradient function and weight evolution with deterministic perturbation, was discussed. In this regard, the regression and character recognition problems were considered. The simulation results, in terms of the convergence rate and global convergence, showed that the integrated approach always outperformed other traditional methods.",

author = "Sin-Chun Ng and Chi-Chung Cheung and Shu-Hung Leung",

year = "2003",

month = mar,

day = "6",

doi = "10.1049/el:20030277",

language = "English",

volume = "39",

pages = "447--448",

journal = "Electronics Letters",

issn = "0013-5194",

publisher = "John Wiley & Sons",

number = "5",

}

TY - JOUR

T1 - Integration of magnified gradient function and weight evolution with deterministic perturbation into back-propagation

AU - Ng, Sin-Chun

AU - Cheung, Chi-Chung

AU - Leung, Shu-Hung

PY - 2003/3/6

Y1 - 2003/3/6

N2 - An approach for the improved performance of back-propagation (BP) learning systems, based on the integration of magnified gradient function and weight evolution with deterministic perturbation, was discussed. In this regard, the regression and character recognition problems were considered. The simulation results, in terms of the convergence rate and global convergence, showed that the integrated approach always outperformed other traditional methods.

AB - An approach for the improved performance of back-propagation (BP) learning systems, based on the integration of magnified gradient function and weight evolution with deterministic perturbation, was discussed. In this regard, the regression and character recognition problems were considered. The simulation results, in terms of the convergence rate and global convergence, showed that the integrated approach always outperformed other traditional methods.

UR - http://www.scopus.com/inward/record.url?scp=0037422119&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-0037422119&origin=recordpage

U2 - 10.1049/el:20030277

DO - 10.1049/el:20030277

M3 - RGC 21 - Publication in refereed journal

SN - 0013-5194

VL - 39

SP - 447

EP - 448

JO - Electronics Letters

JF - Electronics Letters

IS - 5

ER -

Integration of magnified gradient function and weight evolution with deterministic perturbation into back-propagation

Abstract

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this