No Cover Image

Journal article 540 views 194 downloads

Toward deep neural networks: Mirror extreme learning machines for pattern classification

Bolin Liao, Chuan Ma, Meiling Liao, Shuai Li Orcid Logo, Zhiguan Huang

Filomat, Volume: 34, Issue: 15, Pages: 4985 - 4996

Swansea University Author: Shuai Li Orcid Logo

Check full text

DOI (Published version): 10.2298/fil2015985l

Abstract

In this paper, a novel type of feed-forward neural network with a simple structure is proposed and investigated for pattern classification. Because the novel type of forward neural network’s parameter setting is mirrored with those of the Extreme Learning Machine (ELM), it is termed the mirror extre...

Full description

Published in: Filomat
ISSN: 0354-5180 2406-0933
Published: National Library of Serbia 2020
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa56705
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2021-04-20T08:01:29Z
last_indexed 2021-05-25T03:27:41Z
id cronfa56705
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2021-05-24T12:48:13.2221445</datestamp><bib-version>v2</bib-version><id>56705</id><entry>2021-04-20</entry><title>Toward deep neural networks: Mirror extreme learning machines for pattern classification</title><swanseaauthors><author><sid>42ff9eed09bcd109fbbe484a0f99a8a8</sid><ORCID>0000-0001-8316-5289</ORCID><firstname>Shuai</firstname><surname>Li</surname><name>Shuai Li</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2021-04-20</date><deptcode>MECH</deptcode><abstract>In this paper, a novel type of feed-forward neural network with a simple structure is proposed and investigated for pattern classification. Because the novel type of forward neural network&#x2019;s parameter setting is mirrored with those of the Extreme Learning Machine (ELM), it is termed the mirror extreme learning machine (MELM). For the MELM, the input weights are determined by the pseudoinverse method analytically, while the output weights are generated randomly, which are completely different from the conventional ELM. Besides, a growing method is adopted to obtain the optimal hidden-layer structure. Finally, to evaluate the performance of the proposed MELM, abundant comparative experiments based on different real-world classification datasets are performed. Experimental results validate the high classification accuracy and good generalization performance of the proposed neural network with a simple structure in pattern classification.</abstract><type>Journal Article</type><journal>Filomat</journal><volume>34</volume><journalNumber>15</journalNumber><paginationStart>4985</paginationStart><paginationEnd>4996</paginationEnd><publisher>National Library of Serbia</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0354-5180</issnPrint><issnElectronic>2406-0933</issnElectronic><keywords>Mirror extreme learning machine (MELM), Weights determination, Pseudoinverse, Pattern classification, Classification datasets</keywords><publishedDay>1</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2020</publishedYear><publishedDate>2020-01-01</publishedDate><doi>10.2298/fil2015985l</doi><url/><notes/><college>COLLEGE NANME</college><department>Mechanical Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MECH</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2021-05-24T12:48:13.2221445</lastEdited><Created>2021-04-20T08:57:58.5204546</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering</level></path><authors><author><firstname>Bolin</firstname><surname>Liao</surname><order>1</order></author><author><firstname>Chuan</firstname><surname>Ma</surname><order>2</order></author><author><firstname>Meiling</firstname><surname>Liao</surname><order>3</order></author><author><firstname>Shuai</firstname><surname>Li</surname><orcid>0000-0001-8316-5289</orcid><order>4</order></author><author><firstname>Zhiguan</firstname><surname>Huang</surname><order>5</order></author></authors><documents><document><filename>56705__19700__49f85a10ef1f4aa08cbe8a3681d94afe.pdf</filename><originalFilename>56705.pdf</originalFilename><uploaded>2021-04-20T09:00:29.5789408</uploaded><type>Output</type><contentLength>319698</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><copyrightCorrect>false</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2021-05-24T12:48:13.2221445 v2 56705 2021-04-20 Toward deep neural networks: Mirror extreme learning machines for pattern classification 42ff9eed09bcd109fbbe484a0f99a8a8 0000-0001-8316-5289 Shuai Li Shuai Li true false 2021-04-20 MECH In this paper, a novel type of feed-forward neural network with a simple structure is proposed and investigated for pattern classification. Because the novel type of forward neural network’s parameter setting is mirrored with those of the Extreme Learning Machine (ELM), it is termed the mirror extreme learning machine (MELM). For the MELM, the input weights are determined by the pseudoinverse method analytically, while the output weights are generated randomly, which are completely different from the conventional ELM. Besides, a growing method is adopted to obtain the optimal hidden-layer structure. Finally, to evaluate the performance of the proposed MELM, abundant comparative experiments based on different real-world classification datasets are performed. Experimental results validate the high classification accuracy and good generalization performance of the proposed neural network with a simple structure in pattern classification. Journal Article Filomat 34 15 4985 4996 National Library of Serbia 0354-5180 2406-0933 Mirror extreme learning machine (MELM), Weights determination, Pseudoinverse, Pattern classification, Classification datasets 1 1 2020 2020-01-01 10.2298/fil2015985l COLLEGE NANME Mechanical Engineering COLLEGE CODE MECH Swansea University 2021-05-24T12:48:13.2221445 2021-04-20T08:57:58.5204546 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering Bolin Liao 1 Chuan Ma 2 Meiling Liao 3 Shuai Li 0000-0001-8316-5289 4 Zhiguan Huang 5 56705__19700__49f85a10ef1f4aa08cbe8a3681d94afe.pdf 56705.pdf 2021-04-20T09:00:29.5789408 Output 319698 application/pdf Version of Record true false eng
title Toward deep neural networks: Mirror extreme learning machines for pattern classification
spellingShingle Toward deep neural networks: Mirror extreme learning machines for pattern classification
Shuai Li
title_short Toward deep neural networks: Mirror extreme learning machines for pattern classification
title_full Toward deep neural networks: Mirror extreme learning machines for pattern classification
title_fullStr Toward deep neural networks: Mirror extreme learning machines for pattern classification
title_full_unstemmed Toward deep neural networks: Mirror extreme learning machines for pattern classification
title_sort Toward deep neural networks: Mirror extreme learning machines for pattern classification
author_id_str_mv 42ff9eed09bcd109fbbe484a0f99a8a8
author_id_fullname_str_mv 42ff9eed09bcd109fbbe484a0f99a8a8_***_Shuai Li
author Shuai Li
author2 Bolin Liao
Chuan Ma
Meiling Liao
Shuai Li
Zhiguan Huang
format Journal article
container_title Filomat
container_volume 34
container_issue 15
container_start_page 4985
publishDate 2020
institution Swansea University
issn 0354-5180
2406-0933
doi_str_mv 10.2298/fil2015985l
publisher National Library of Serbia
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering
document_store_str 1
active_str 0
description In this paper, a novel type of feed-forward neural network with a simple structure is proposed and investigated for pattern classification. Because the novel type of forward neural network’s parameter setting is mirrored with those of the Extreme Learning Machine (ELM), it is termed the mirror extreme learning machine (MELM). For the MELM, the input weights are determined by the pseudoinverse method analytically, while the output weights are generated randomly, which are completely different from the conventional ELM. Besides, a growing method is adopted to obtain the optimal hidden-layer structure. Finally, to evaluate the performance of the proposed MELM, abundant comparative experiments based on different real-world classification datasets are performed. Experimental results validate the high classification accuracy and good generalization performance of the proposed neural network with a simple structure in pattern classification.
published_date 2020-01-01T04:11:52Z
_version_ 1763753813131395072
score 11.036334

Similar Items