Noise resistant training for extreme learning machine

Yik Lam LUI; Hiu Tung WONG; Chi-Sing LEUNG; Sam KWONG

doi:10.1007/978-3-319-59081-3_31

Noise resistant training for extreme learning machine

Yik Lam LUI, Hiu Tung WONG, Chi-Sing LEUNG, Sam KWONG

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

1 Citation (Scopus)

Abstract

The extreme learning machine (ELM) concept provides some effective training algorithms to construct single hidden layer feedforward networks (SHLFNs). However, the conventional ELM algorithms were designed for the noiseless situation only, in which the outputs of the hidden nodes are not contaminated by noise. This paper presents two noise-resistant training algorithms, namely noise-resistant incremental ELM (NRI-ELM) and noise-resistant convex incremental ELM (NRCI-ELM). For NRI-ELM, its noise-resistant ability is better than that of the conventional incremented ELM algorithms. To further enhance the noise resistant ability, the NRCI-ELM algorithm is proposed. The convergent properties of the two proposed noise resistant algorithms are also presented.

Original language	English
Title of host publication	Advances in Neural Networks: ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017, Proceedings, Part II
Editors	Fengyu CONG, Andrew LEUNG, Qinglai WEI
Publisher	Springer, Cham
Pages	257-265
Number of pages	9
ISBN (Electronic)	9783319590813
ISBN (Print)	9783319590806
DOIs	https://doi.org/10.1007/978-3-319-59081-3_31
Publication status	Published - 2017
Externally published	Yes
Event	14th International Symposium on Neural Networks (ISNN 2017) - Hokkaido University, Sapporo, Japan Duration: 21 Jun 2017 → 26 Jun 2017

Conference

Conference	14th International Symposium on Neural Networks (ISNN 2017)
Country/Territory	Japan
City	Sapporo
Period	21/06/17 → 26/06/17

Funding

The work was supported by a research grant from the Government of the Hong Kong Special Administrative Region (CityU 11259516).

Keywords

Extreme learning machines
Incremental algorithm
Node noise

Access to Document

10.1007/978-3-319-59081-3_31

Cite this

LUI, Y. L., WONG, H. T., LEUNG, C.-S., & KWONG, S. (2017). Noise resistant training for extreme learning machine. In F. CONG, A. LEUNG, & Q. WEI (Eds.), Advances in Neural Networks: ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017, Proceedings, Part II (pp. 257-265). Springer, Cham. https://doi.org/10.1007/978-3-319-59081-3_31

@inproceedings{6dbc15ee6fc84f8792dfc04b4eea72b3,

title = "Noise resistant training for extreme learning machine",

abstract = "The extreme learning machine (ELM) concept provides some effective training algorithms to construct single hidden layer feedforward networks (SHLFNs). However, the conventional ELM algorithms were designed for the noiseless situation only, in which the outputs of the hidden nodes are not contaminated by noise. This paper presents two noise-resistant training algorithms, namely noise-resistant incremental ELM (NRI-ELM) and noise-resistant convex incremental ELM (NRCI-ELM). For NRI-ELM, its noise-resistant ability is better than that of the conventional incremented ELM algorithms. To further enhance the noise resistant ability, the NRCI-ELM algorithm is proposed. The convergent properties of the two proposed noise resistant algorithms are also presented.",

keywords = "Extreme learning machines, Incremental algorithm, Node noise",

author = "LUI, {Yik Lam} and WONG, {Hiu Tung} and Chi-Sing LEUNG and Sam KWONG",

year = "2017",

doi = "10.1007/978-3-319-59081-3_31",

language = "English",

isbn = "9783319590806",

pages = "257--265",

editor = "Fengyu CONG and Andrew LEUNG and Qinglai WEI",

booktitle = "Advances in Neural Networks: ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017, Proceedings, Part II",

publisher = "Springer, Cham",

note = "14th International Symposium on Neural Networks (ISNN 2017) ; Conference date: 21-06-2017 Through 26-06-2017",

}

LUI, YL, WONG, HT, LEUNG, C-S & KWONG, S 2017, Noise resistant training for extreme learning machine. in F CONG, A LEUNG & Q WEI (eds), Advances in Neural Networks: ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017, Proceedings, Part II. Springer, Cham, pp. 257-265, 14th International Symposium on Neural Networks (ISNN 2017), Sapporo, Japan, 21/06/17. https://doi.org/10.1007/978-3-319-59081-3_31

Noise resistant training for extreme learning machine. / LUI, Yik Lam; WONG, Hiu Tung; LEUNG, Chi-Sing et al.
Advances in Neural Networks: ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017, Proceedings, Part II. ed. / Fengyu CONG; Andrew LEUNG; Qinglai WEI. Springer, Cham, 2017. p. 257-265.

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

TY - GEN

T1 - Noise resistant training for extreme learning machine

AU - LUI, Yik Lam

AU - WONG, Hiu Tung

AU - LEUNG, Chi-Sing

AU - KWONG, Sam

PY - 2017

Y1 - 2017

N2 - The extreme learning machine (ELM) concept provides some effective training algorithms to construct single hidden layer feedforward networks (SHLFNs). However, the conventional ELM algorithms were designed for the noiseless situation only, in which the outputs of the hidden nodes are not contaminated by noise. This paper presents two noise-resistant training algorithms, namely noise-resistant incremental ELM (NRI-ELM) and noise-resistant convex incremental ELM (NRCI-ELM). For NRI-ELM, its noise-resistant ability is better than that of the conventional incremented ELM algorithms. To further enhance the noise resistant ability, the NRCI-ELM algorithm is proposed. The convergent properties of the two proposed noise resistant algorithms are also presented.

AB - The extreme learning machine (ELM) concept provides some effective training algorithms to construct single hidden layer feedforward networks (SHLFNs). However, the conventional ELM algorithms were designed for the noiseless situation only, in which the outputs of the hidden nodes are not contaminated by noise. This paper presents two noise-resistant training algorithms, namely noise-resistant incremental ELM (NRI-ELM) and noise-resistant convex incremental ELM (NRCI-ELM). For NRI-ELM, its noise-resistant ability is better than that of the conventional incremented ELM algorithms. To further enhance the noise resistant ability, the NRCI-ELM algorithm is proposed. The convergent properties of the two proposed noise resistant algorithms are also presented.

KW - Extreme learning machines

KW - Incremental algorithm

KW - Node noise

UR - http://www.scopus.com/inward/record.url?scp=85021733548&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-59081-3_31

DO - 10.1007/978-3-319-59081-3_31

M3 - Conference paper (refereed)

SN - 9783319590806

SP - 257

EP - 265

BT - Advances in Neural Networks: ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017, Proceedings, Part II

A2 - CONG, Fengyu

A2 - LEUNG, Andrew

A2 - WEI, Qinglai

PB - Springer, Cham

T2 - 14th International Symposium on Neural Networks (ISNN 2017)

Y2 - 21 June 2017 through 26 June 2017

ER -

Noise resistant training for extreme learning machine

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this