A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise

Muideen Adegoke; Yuqi Xiao; Chi-Sing Leung; Kwok Wa Leung

doi:10.1007/978-3-031-00828-3_7

A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise

Muideen Adegoke, Yuqi Xiao, Chi-Sing Leung^*, Kwok Wa Leung

^*Corresponding author for this work

Department of Electrical Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 12 - Chapter in an edited book (Author) › peer-review

Abstract

Although the extreme learning machine (ELM) technique is an efficient and effective neural approach, there are still some downsides in the traditional ELM technique. When there are some outlier training samples, the trained neural network is usually with poor performance. Another issue is that when there are some noise and faults in the trained network, the performance of the trained network is also poor. This paper looks into the ELM technique under multiple imperfections, including outlier training samples, weight noise and node faults. This paper first identifies a regularization term for handling weight noise and node faults. To handle outlier training samples, the maximum correntropy criterion (MCC) concept is used in the objective function. A learning algorithm, namely, robust fault aware ELM algorithm (RFAELM), for faulty networks is then proposed. Simulation results show that the performance of the proposed algorithm is much better than that of two state-of-art robust algorithms.

Original language	English
Title of host publication	Recent Advances in Soft Computing and Data Mining
Subtitle of host publication	Proceedings of the Fifth International Conference on Soft Computing and Data Mining (SCDM 2022), May 30-31, 2022
Editors	Rozaida Ghazali, Nazri Mohd Nawi, Mustafa Mat Deris, Jemal H. Abawajy, Nureize Arbaiy
Place of Publication	Cham
Publisher	Springer
Pages	69-78
ISBN (Electronic)	978-3-031-00828-3
ISBN (Print)	9783031008276
DOIs	https://doi.org/10.1007/978-3-031-00828-3_7
Publication status	Published - 2022
Event	5th International Conference on Soft Computing and Data Mining (SCDM 2022) - Universiti Tun Hussein Onn Malaysia (Virtual), Johor, Malaysia Duration: 30 May 2022 → 31 May 2022 https://scdm.uthm.edu.my/scdm2022/

Publication series

Name	Lecture Notes in Networks and Systems
Volume	457
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	5th International Conference on Soft Computing and Data Mining (SCDM 2022)
Country/Territory	Malaysia
City	Johor
Period	30/05/22 → 31/05/22
Internet address	https://scdm.uthm.edu.my/scdm2022/

Bibliographical note

Information for this record is supplemented by the author(s) concerned.

Research Keywords

Node fault
Outlier samples
Weight noise

Access Output

10.1007/978-3-031-00828-3_7

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Adegoke, M., Xiao, Y., Leung, C.-S., & Leung, K. W. (2022). A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise. In R. Ghazali, N. M. Nawi, M. M. Deris, J. H. Abawajy, & N. Arbaiy (Eds.), Recent Advances in Soft Computing and Data Mining: Proceedings of the Fifth International Conference on Soft Computing and Data Mining (SCDM 2022), May 30-31, 2022 (pp. 69-78). (Lecture Notes in Networks and Systems; Vol. 457). Springer . https://doi.org/10.1007/978-3-031-00828-3_7

Adegoke, Muideen ; Xiao, Yuqi ; Leung, Chi-Sing et al. / A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise. Recent Advances in Soft Computing and Data Mining: Proceedings of the Fifth International Conference on Soft Computing and Data Mining (SCDM 2022), May 30-31, 2022. editor / Rozaida Ghazali ; Nazri Mohd Nawi ; Mustafa Mat Deris ; Jemal H. Abawajy ; Nureize Arbaiy. Cham : Springer , 2022. pp. 69-78 (Lecture Notes in Networks and Systems).

@inbook{43af13913c9649c48d206e7b0abdd892,

title = "A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise",

abstract = "Although the extreme learning machine (ELM) technique is an efficient and effective neural approach, there are still some downsides in the traditional ELM technique. When there are some outlier training samples, the trained neural network is usually with poor performance. Another issue is that when there are some noise and faults in the trained network, the performance of the trained network is also poor. This paper looks into the ELM technique under multiple imperfections, including outlier training samples, weight noise and node faults. This paper first identifies a regularization term for handling weight noise and node faults. To handle outlier training samples, the maximum correntropy criterion (MCC) concept is used in the objective function. A learning algorithm, namely, robust fault aware ELM algorithm (RFAELM), for faulty networks is then proposed. Simulation results show that the performance of the proposed algorithm is much better than that of two state-of-art robust algorithms.",

keywords = "Node fault, Outlier samples, Weight noise",

author = "Muideen Adegoke and Yuqi Xiao and Chi-Sing Leung and Leung, {Kwok Wa}",

note = "Information for this record is supplemented by the author(s) concerned.; 5th International Conference on Soft Computing and Data Mining (SCDM 2022) ; Conference date: 30-05-2022 Through 31-05-2022",

year = "2022",

doi = "10.1007/978-3-031-00828-3_7",

language = "English",

isbn = "9783031008276",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer ",

pages = "69--78",

editor = "Rozaida Ghazali and Nawi, {Nazri Mohd} and Deris, {Mustafa Mat} and Abawajy, {Jemal H.} and Nureize Arbaiy",

booktitle = "Recent Advances in Soft Computing and Data Mining",

url = "https://scdm.uthm.edu.my/scdm2022/",

}

Adegoke, M , Xiao, Y , Leung, C-S & Leung, KW 2022, A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise. in R Ghazali, NM Nawi, MM Deris, JH Abawajy & N Arbaiy (eds), Recent Advances in Soft Computing and Data Mining: Proceedings of the Fifth International Conference on Soft Computing and Data Mining (SCDM 2022), May 30-31, 2022. Lecture Notes in Networks and Systems, vol. 457, Springer , Cham, pp. 69-78, 5th International Conference on Soft Computing and Data Mining (SCDM 2022), Johor, Malaysia, 30/05/22. https://doi.org/10.1007/978-3-031-00828-3_7

A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise. / Adegoke, Muideen ; Xiao, Yuqi ; Leung, Chi-Sing et al.
Recent Advances in Soft Computing and Data Mining: Proceedings of the Fifth International Conference on Soft Computing and Data Mining (SCDM 2022), May 30-31, 2022. ed. / Rozaida Ghazali; Nazri Mohd Nawi; Mustafa Mat Deris; Jemal H. Abawajy; Nureize Arbaiy. Cham: Springer , 2022. p. 69-78 (Lecture Notes in Networks and Systems; Vol. 457).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 12 - Chapter in an edited book (Author) › peer-review

TY - CHAP

T1 - A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise

AU - Adegoke, Muideen

AU - Xiao, Yuqi

AU - Leung, Chi-Sing

AU - Leung, Kwok Wa

N1 - Information for this record is supplemented by the author(s) concerned.

PY - 2022

Y1 - 2022

N2 - Although the extreme learning machine (ELM) technique is an efficient and effective neural approach, there are still some downsides in the traditional ELM technique. When there are some outlier training samples, the trained neural network is usually with poor performance. Another issue is that when there are some noise and faults in the trained network, the performance of the trained network is also poor. This paper looks into the ELM technique under multiple imperfections, including outlier training samples, weight noise and node faults. This paper first identifies a regularization term for handling weight noise and node faults. To handle outlier training samples, the maximum correntropy criterion (MCC) concept is used in the objective function. A learning algorithm, namely, robust fault aware ELM algorithm (RFAELM), for faulty networks is then proposed. Simulation results show that the performance of the proposed algorithm is much better than that of two state-of-art robust algorithms.

AB - Although the extreme learning machine (ELM) technique is an efficient and effective neural approach, there are still some downsides in the traditional ELM technique. When there are some outlier training samples, the trained neural network is usually with poor performance. Another issue is that when there are some noise and faults in the trained network, the performance of the trained network is also poor. This paper looks into the ELM technique under multiple imperfections, including outlier training samples, weight noise and node faults. This paper first identifies a regularization term for handling weight noise and node faults. To handle outlier training samples, the maximum correntropy criterion (MCC) concept is used in the objective function. A learning algorithm, namely, robust fault aware ELM algorithm (RFAELM), for faulty networks is then proposed. Simulation results show that the performance of the proposed algorithm is much better than that of two state-of-art robust algorithms.

KW - Node fault

KW - Outlier samples

KW - Weight noise

UR - http://www.scopus.com/inward/record.url?scp=85130340183&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85130340183&origin=recordpage

U2 - 10.1007/978-3-031-00828-3_7

DO - 10.1007/978-3-031-00828-3_7

M3 - RGC 12 - Chapter in an edited book (Author)

SN - 9783031008276

T3 - Lecture Notes in Networks and Systems

SP - 69

EP - 78

BT - Recent Advances in Soft Computing and Data Mining

A2 - Ghazali, Rozaida

A2 - Nawi, Nazri Mohd

A2 - Deris, Mustafa Mat

A2 - Abawajy, Jemal H.

A2 - Arbaiy, Nureize

PB - Springer

CY - Cham

T2 - 5th International Conference on Soft Computing and Data Mining (SCDM 2022)

Y2 - 30 May 2022 through 31 May 2022

ER -

Adegoke M , Xiao Y , Leung CS , Leung KW. A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise. In Ghazali R, Nawi NM, Deris MM, Abawajy JH, Arbaiy N, editors, Recent Advances in Soft Computing and Data Mining: Proceedings of the Fifth International Conference on Soft Computing and Data Mining (SCDM 2022), May 30-31, 2022. Cham: Springer . 2022. p. 69-78. (Lecture Notes in Networks and Systems). Epub 2022 May 4. doi: 10.1007/978-3-031-00828-3_7

A Robust ELM Algorithm for Compensating the Effect of Node Fault and Weight Noise

Abstract

Publication series

Conference

Bibliographical note

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this