No Cover Image

Journal article 711 views 796 downloads

Graph Deep Learning: State of the Art and Challenges

Stavros Georgousis, Michael Kenning, Xianghua Xie Orcid Logo

IEEE Access, Volume: 9, Pages: 22106 - 22140

Swansea University Authors: Michael Kenning, Xianghua Xie Orcid Logo

  • 56088.pdf

    PDF | Version of Record

    Released under the terms of a Creative Commons Attribution 4.0 License

    Download (5.94MB)

Check full text

DOI (Published version): 10.1109/access.2021.3055280

Abstract

The last half-decade has seen a surge in deep learning research on irregular domains and efforts to extend convolutional neural networks (CNNs) to work on irregularly structured data. The graph has emerged as a particularly useful geometrical object in deep learning, able to represent a variety of i...

Full description

Published in: IEEE Access
ISSN: 2169-3536
Published: Institute of Electrical and Electronics Engineers (IEEE) 2021
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa56088
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2021-01-21T12:01:12Z
last_indexed 2021-02-23T04:20:07Z
id cronfa56088
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2021-02-22T13:05:23.8665456</datestamp><bib-version>v2</bib-version><id>56088</id><entry>2021-01-21</entry><title>Graph Deep Learning: State of the Art and Challenges</title><swanseaauthors><author><sid>3fcab7bac19385191914aa7e98b88e07</sid><firstname>Michael</firstname><surname>Kenning</surname><name>Michael Kenning</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>b334d40963c7a2f435f06d2c26c74e11</sid><ORCID>0000-0002-2701-8660</ORCID><firstname>Xianghua</firstname><surname>Xie</surname><name>Xianghua Xie</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2021-01-21</date><deptcode>MRKT</deptcode><abstract>The last half-decade has seen a surge in deep learning research on irregular domains and efforts to extend convolutional neural networks (CNNs) to work on irregularly structured data. The graph has emerged as a particularly useful geometrical object in deep learning, able to represent a variety of irregular domains well. Graphs can represent various complex systems, from molecular structure, to computer and social and traffic networks. Consequent on the extension of CNNs to graphs, a great amount of research has been published that improves the inferential power and computational efficiency of graph- based convolutional neural networks (GCNNs).The research is incipient, however, and our understanding is relatively rudimentary. The majority of GCNNs are designed to operate with certain properties. In this survey we review of the state of graph representation learning from the perspective of deep learning. We consider challenges in graph deep learning that have been neglected in the majority of work, largely because of the numerous theoretical difficulties they present. We identify four major challenges in graph deep learning: dynamic and evolving graphs, learning with edge signals and information, graph estimation, and the generalization of graph models. For each problem we discuss the theoretical and practical issues, survey the relevant research, while highlighting the limitations of the state of the art. Advances on these challenges would permit GCNNs to be applied to wider range of domains, in situations where graph models have previously been limited owing to the obstructions to applying a model owing to the domains&#x2019; natures.</abstract><type>Journal Article</type><journal>IEEE Access</journal><volume>9</volume><journalNumber/><paginationStart>22106</paginationStart><paginationEnd>22140</paginationEnd><publisher>Institute of Electrical and Electronics Engineers (IEEE)</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>2169-3536</issnElectronic><keywords/><publishedDay>28</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2021</publishedYear><publishedDate>2021-01-28</publishedDate><doi>10.1109/access.2021.3055280</doi><url/><notes/><college>COLLEGE NANME</college><department>Marketing</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MRKT</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2021-02-22T13:05:23.8665456</lastEdited><Created>2021-01-21T09:33:05.9099983</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Stavros</firstname><surname>Georgousis</surname><order>1</order></author><author><firstname>Michael</firstname><surname>Kenning</surname><order>2</order></author><author><firstname>Xianghua</firstname><surname>Xie</surname><orcid>0000-0002-2701-8660</orcid><order>3</order></author></authors><documents><document><filename>56088__19344__ab0dda3259d940c283f1cd378fcdd7a1.pdf</filename><originalFilename>56088.pdf</originalFilename><uploaded>2021-02-22T13:02:18.1896336</uploaded><type>Output</type><contentLength>6227385</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>Released under the terms of a Creative Commons Attribution 4.0 License</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2021-02-22T13:05:23.8665456 v2 56088 2021-01-21 Graph Deep Learning: State of the Art and Challenges 3fcab7bac19385191914aa7e98b88e07 Michael Kenning Michael Kenning true false b334d40963c7a2f435f06d2c26c74e11 0000-0002-2701-8660 Xianghua Xie Xianghua Xie true false 2021-01-21 MRKT The last half-decade has seen a surge in deep learning research on irregular domains and efforts to extend convolutional neural networks (CNNs) to work on irregularly structured data. The graph has emerged as a particularly useful geometrical object in deep learning, able to represent a variety of irregular domains well. Graphs can represent various complex systems, from molecular structure, to computer and social and traffic networks. Consequent on the extension of CNNs to graphs, a great amount of research has been published that improves the inferential power and computational efficiency of graph- based convolutional neural networks (GCNNs).The research is incipient, however, and our understanding is relatively rudimentary. The majority of GCNNs are designed to operate with certain properties. In this survey we review of the state of graph representation learning from the perspective of deep learning. We consider challenges in graph deep learning that have been neglected in the majority of work, largely because of the numerous theoretical difficulties they present. We identify four major challenges in graph deep learning: dynamic and evolving graphs, learning with edge signals and information, graph estimation, and the generalization of graph models. For each problem we discuss the theoretical and practical issues, survey the relevant research, while highlighting the limitations of the state of the art. Advances on these challenges would permit GCNNs to be applied to wider range of domains, in situations where graph models have previously been limited owing to the obstructions to applying a model owing to the domains’ natures. Journal Article IEEE Access 9 22106 22140 Institute of Electrical and Electronics Engineers (IEEE) 2169-3536 28 1 2021 2021-01-28 10.1109/access.2021.3055280 COLLEGE NANME Marketing COLLEGE CODE MRKT Swansea University 2021-02-22T13:05:23.8665456 2021-01-21T09:33:05.9099983 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Stavros Georgousis 1 Michael Kenning 2 Xianghua Xie 0000-0002-2701-8660 3 56088__19344__ab0dda3259d940c283f1cd378fcdd7a1.pdf 56088.pdf 2021-02-22T13:02:18.1896336 Output 6227385 application/pdf Version of Record true Released under the terms of a Creative Commons Attribution 4.0 License true eng https://creativecommons.org/licenses/by/4.0/
title Graph Deep Learning: State of the Art and Challenges
spellingShingle Graph Deep Learning: State of the Art and Challenges
Michael Kenning
Xianghua Xie
title_short Graph Deep Learning: State of the Art and Challenges
title_full Graph Deep Learning: State of the Art and Challenges
title_fullStr Graph Deep Learning: State of the Art and Challenges
title_full_unstemmed Graph Deep Learning: State of the Art and Challenges
title_sort Graph Deep Learning: State of the Art and Challenges
author_id_str_mv 3fcab7bac19385191914aa7e98b88e07
b334d40963c7a2f435f06d2c26c74e11
author_id_fullname_str_mv 3fcab7bac19385191914aa7e98b88e07_***_Michael Kenning
b334d40963c7a2f435f06d2c26c74e11_***_Xianghua Xie
author Michael Kenning
Xianghua Xie
author2 Stavros Georgousis
Michael Kenning
Xianghua Xie
format Journal article
container_title IEEE Access
container_volume 9
container_start_page 22106
publishDate 2021
institution Swansea University
issn 2169-3536
doi_str_mv 10.1109/access.2021.3055280
publisher Institute of Electrical and Electronics Engineers (IEEE)
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 Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description The last half-decade has seen a surge in deep learning research on irregular domains and efforts to extend convolutional neural networks (CNNs) to work on irregularly structured data. The graph has emerged as a particularly useful geometrical object in deep learning, able to represent a variety of irregular domains well. Graphs can represent various complex systems, from molecular structure, to computer and social and traffic networks. Consequent on the extension of CNNs to graphs, a great amount of research has been published that improves the inferential power and computational efficiency of graph- based convolutional neural networks (GCNNs).The research is incipient, however, and our understanding is relatively rudimentary. The majority of GCNNs are designed to operate with certain properties. In this survey we review of the state of graph representation learning from the perspective of deep learning. We consider challenges in graph deep learning that have been neglected in the majority of work, largely because of the numerous theoretical difficulties they present. We identify four major challenges in graph deep learning: dynamic and evolving graphs, learning with edge signals and information, graph estimation, and the generalization of graph models. For each problem we discuss the theoretical and practical issues, survey the relevant research, while highlighting the limitations of the state of the art. Advances on these challenges would permit GCNNs to be applied to wider range of domains, in situations where graph models have previously been limited owing to the obstructions to applying a model owing to the domains’ natures.
published_date 2021-01-28T04:10:46Z
_version_ 1763753744654139392
score 11.037603

Similar Items