Multi-criteria decision making based architecture selection for single-hidden layer feedforward neural networks

Ran WANG*, Haoran XIE, Jiqiang FENG, Fu Lee WANG, Chen XU

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)

1 Scopus Citations

Abstract

Architecture selection is a fundamental problem in artificial neural networks, which could be treated as a decision making process that evaluates, ranks, and makes choices from a set of network structures. Traditional methods evaluate a network structure by designing a criterion based on a validation model or an error bound model. On one hand, the time complexity of a validation model is usually high; on the other hand, different validation models or error bound models may lead to different (even conflicting) results, which post challenges to the traditional single criterion-based architecture selection methods. In the area of decision making, many problems employed multiple criteria since the performance is better than using a single criterion. In this paper, we propose a multi-criteria decision making based architecture selection algorithm for single-hidden layer feedforward neural networks trained by extreme learning machine. Two criteria are incorporated into the selection process, i.e., training accuracy and the Q-value estimated by the localized generalization error model. The training accuracy reflects the capability of the model on correctly categorizing the known samples, and the Q-value estimated by localized generalization error model reflects the size of the neighbourhood of training samples in which the model can predict unseen samples with confidence. By achieving a trade-off between these two criteria, a new architecture selection algorithm is proposed. Experimental comparisons demonstrate the feasibility and effectiveness of the proposed method.

Original languageEnglish
Pages (from-to)655-666
Number of pages12
JournalInternational Journal of Machine Learning and Cybernetics
Volume10
Issue number4
Early online date18 Nov 2017
DOIs
Publication statusPublished - 2 Apr 2019
Externally publishedYes

Fingerprint

Feedforward neural networks
Decision making
Learning systems
Neural networks

Bibliographical note

This work was supported in part by the National Natural Science Foundation of China under Grant 61772344, Grant 61402460, Grant 61732011, and Grant 61472257, in part by the Natural Science Foundation of SZU under Grant 2017060, in part by the Guangdong Provincial Science and Technology Plan Project under Grant 2013B040403005, in part by the HD Video R&D Platform for Intelligent Analysis and Processing in Guangdong Engineering Technology Research Centre of Colleges and Universities under Grant GCZX-A1409, in part by the Internal Research Grant (RG 66/2016-2017) of The Education University of Hong Kong, and in part by a grant from Research Grants Council of Hong Kong Special Administrative Region, China (UGC/FDS11/E04/16).

Keywords

  • Architecture selection
  • Extreme learning machine
  • Localized generalization error model
  • Multi-criteria decision making

Cite this

@article{9ebe7c13f26d46d3a32a59511ad67246,
title = "Multi-criteria decision making based architecture selection for single-hidden layer feedforward neural networks",
abstract = "Architecture selection is a fundamental problem in artificial neural networks, which could be treated as a decision making process that evaluates, ranks, and makes choices from a set of network structures. Traditional methods evaluate a network structure by designing a criterion based on a validation model or an error bound model. On one hand, the time complexity of a validation model is usually high; on the other hand, different validation models or error bound models may lead to different (even conflicting) results, which post challenges to the traditional single criterion-based architecture selection methods. In the area of decision making, many problems employed multiple criteria since the performance is better than using a single criterion. In this paper, we propose a multi-criteria decision making based architecture selection algorithm for single-hidden layer feedforward neural networks trained by extreme learning machine. Two criteria are incorporated into the selection process, i.e., training accuracy and the Q-value estimated by the localized generalization error model. The training accuracy reflects the capability of the model on correctly categorizing the known samples, and the Q-value estimated by localized generalization error model reflects the size of the neighbourhood of training samples in which the model can predict unseen samples with confidence. By achieving a trade-off between these two criteria, a new architecture selection algorithm is proposed. Experimental comparisons demonstrate the feasibility and effectiveness of the proposed method.",
keywords = "Architecture selection, Extreme learning machine, Localized generalization error model, Multi-criteria decision making",
author = "Ran WANG and Haoran XIE and Jiqiang FENG and WANG, {Fu Lee} and Chen XU",
note = "This work was supported in part by the National Natural Science Foundation of China under Grant 61772344, Grant 61402460, Grant 61732011, and Grant 61472257, in part by the Natural Science Foundation of SZU under Grant 2017060, in part by the Guangdong Provincial Science and Technology Plan Project under Grant 2013B040403005, in part by the HD Video R&D Platform for Intelligent Analysis and Processing in Guangdong Engineering Technology Research Centre of Colleges and Universities under Grant GCZX-A1409, in part by the Internal Research Grant (RG 66/2016-2017) of The Education University of Hong Kong, and in part by a grant from Research Grants Council of Hong Kong Special Administrative Region, China (UGC/FDS11/E04/16).",
year = "2019",
month = "4",
day = "2",
doi = "10.1007/s13042-017-0746-9",
language = "English",
volume = "10",
pages = "655--666",
journal = "International Journal of Machine Learning and Cybernetics",
issn = "1868-8071",
publisher = "Springer Science + Business Media",
number = "4",

}

Multi-criteria decision making based architecture selection for single-hidden layer feedforward neural networks. / WANG, Ran; XIE, Haoran; FENG, Jiqiang; WANG, Fu Lee; XU, Chen.

In: International Journal of Machine Learning and Cybernetics, Vol. 10, No. 4, 02.04.2019, p. 655-666.

Research output: Journal PublicationsJournal Article (refereed)

TY - JOUR

T1 - Multi-criteria decision making based architecture selection for single-hidden layer feedforward neural networks

AU - WANG, Ran

AU - XIE, Haoran

AU - FENG, Jiqiang

AU - WANG, Fu Lee

AU - XU, Chen

N1 - This work was supported in part by the National Natural Science Foundation of China under Grant 61772344, Grant 61402460, Grant 61732011, and Grant 61472257, in part by the Natural Science Foundation of SZU under Grant 2017060, in part by the Guangdong Provincial Science and Technology Plan Project under Grant 2013B040403005, in part by the HD Video R&D Platform for Intelligent Analysis and Processing in Guangdong Engineering Technology Research Centre of Colleges and Universities under Grant GCZX-A1409, in part by the Internal Research Grant (RG 66/2016-2017) of The Education University of Hong Kong, and in part by a grant from Research Grants Council of Hong Kong Special Administrative Region, China (UGC/FDS11/E04/16).

PY - 2019/4/2

Y1 - 2019/4/2

N2 - Architecture selection is a fundamental problem in artificial neural networks, which could be treated as a decision making process that evaluates, ranks, and makes choices from a set of network structures. Traditional methods evaluate a network structure by designing a criterion based on a validation model or an error bound model. On one hand, the time complexity of a validation model is usually high; on the other hand, different validation models or error bound models may lead to different (even conflicting) results, which post challenges to the traditional single criterion-based architecture selection methods. In the area of decision making, many problems employed multiple criteria since the performance is better than using a single criterion. In this paper, we propose a multi-criteria decision making based architecture selection algorithm for single-hidden layer feedforward neural networks trained by extreme learning machine. Two criteria are incorporated into the selection process, i.e., training accuracy and the Q-value estimated by the localized generalization error model. The training accuracy reflects the capability of the model on correctly categorizing the known samples, and the Q-value estimated by localized generalization error model reflects the size of the neighbourhood of training samples in which the model can predict unseen samples with confidence. By achieving a trade-off between these two criteria, a new architecture selection algorithm is proposed. Experimental comparisons demonstrate the feasibility and effectiveness of the proposed method.

AB - Architecture selection is a fundamental problem in artificial neural networks, which could be treated as a decision making process that evaluates, ranks, and makes choices from a set of network structures. Traditional methods evaluate a network structure by designing a criterion based on a validation model or an error bound model. On one hand, the time complexity of a validation model is usually high; on the other hand, different validation models or error bound models may lead to different (even conflicting) results, which post challenges to the traditional single criterion-based architecture selection methods. In the area of decision making, many problems employed multiple criteria since the performance is better than using a single criterion. In this paper, we propose a multi-criteria decision making based architecture selection algorithm for single-hidden layer feedforward neural networks trained by extreme learning machine. Two criteria are incorporated into the selection process, i.e., training accuracy and the Q-value estimated by the localized generalization error model. The training accuracy reflects the capability of the model on correctly categorizing the known samples, and the Q-value estimated by localized generalization error model reflects the size of the neighbourhood of training samples in which the model can predict unseen samples with confidence. By achieving a trade-off between these two criteria, a new architecture selection algorithm is proposed. Experimental comparisons demonstrate the feasibility and effectiveness of the proposed method.

KW - Architecture selection

KW - Extreme learning machine

KW - Localized generalization error model

KW - Multi-criteria decision making

UR - http://www.scopus.com/inward/record.url?scp=85046580924&partnerID=8YFLogxK

U2 - 10.1007/s13042-017-0746-9

DO - 10.1007/s13042-017-0746-9

M3 - Journal Article (refereed)

AN - SCOPUS:85046580924

VL - 10

SP - 655

EP - 666

JO - International Journal of Machine Learning and Cybernetics

JF - International Journal of Machine Learning and Cybernetics

SN - 1868-8071

IS - 4

ER -