HEAT: Hyperbolic Embedding of Attributed Networks

David McDonald; Shan He

doi:10.1007/978-3-030-62362-3_4

HEAT: Hyperbolic Embedding of Attributed Networks

David McDonald^*, Shan He

^*Corresponding author for this work

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

30 Downloads (Pure)

Abstract

Finding a low dimensional representation of hierarchical, structured data described by a network remains a challenging problem in the machine learning community. An emerging approach is embedding networks into hyperbolic space since it can naturally represent a network’s hierarchical structure. However, existing hyperbolic embedding approaches cannot deal with attributed networks, where nodes are annotated with additional attributes. To overcome this, we propose HEAT (for Hyperbolic Embedding of Attributed Networks). HEAT first extracts training samples from the network that captures both topological and attribute node similarity and then learns a low-dimensional hyperboloid embedding using full Riemannian Stochastic Gradient Descent. We show that HEAT can outperform other network embedding algorithms on several common downstream tasks. As a general network embedding method, HEAT opens the door to hyperbolic manifold learning on a wide range of both attributed and unattributed networks.

Original language	English
Title of host publication	Intelligent Data Engineering and Automated Learning – IDEAL 2020
Subtitle of host publication	21st International Conference, Guimaraes, Portugal, November 4–6, 2020, Proceedings, Part I
Editors	Cesar Analide, Paulo Novais, David Camacho, Hujun Yin
Publisher	Springer
Pages	28-40
Number of pages	13
Edition	1
ISBN (Electronic)	9783030623623
ISBN (Print)	9783030623616
DOIs	https://doi.org/10.1007/978-3-030-62362-3_4
Publication status	Published - 30 Oct 2020
Event	21th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2020 - Guimaraes, Portugal Duration: 4 Nov 2020 → 6 Nov 2020

Publication series

Name	Lecture Notes in Computer Science
Volume	12489
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	21th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2020
Country/Territory	Portugal
City	Guimaraes
Period	4/11/20 → 6/11/20

Bibliographical note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.

Keywords

Network embedding
Hyperbolic embedding
Random walk

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-62362-3_4

McDonaldD2020HEAT
This version of the contribution has been accepted for publication, after peer review (when applicable) but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/10.1007/978-3-030-62362-3_4. Use of this Accepted Version is subject to the publisher’s Accepted Manuscript terms of use: https://www.springernature.com/gp/open-research/policies/accepted-manuscript-terms
Accepted author manuscript, 498 KBLicence: Other (please specify with Rights Statement)

Cite this

McDonald, D., & He, S. (2020). HEAT: Hyperbolic Embedding of Attributed Networks. In C. Analide, P. Novais, D. Camacho, & H. Yin (Eds.), Intelligent Data Engineering and Automated Learning – IDEAL 2020: 21st International Conference, Guimaraes, Portugal, November 4–6, 2020, Proceedings, Part I (1 ed., pp. 28-40). (Lecture Notes in Computer Science; Vol. 12489). Springer. https://doi.org/10.1007/978-3-030-62362-3_4

McDonald, David ; He, Shan. / HEAT : Hyperbolic Embedding of Attributed Networks. Intelligent Data Engineering and Automated Learning – IDEAL 2020: 21st International Conference, Guimaraes, Portugal, November 4–6, 2020, Proceedings, Part I. editor / Cesar Analide ; Paulo Novais ; David Camacho ; Hujun Yin. 1. ed. Springer, 2020. pp. 28-40 (Lecture Notes in Computer Science).

@inproceedings{220e6a0982b2427f877e5f1388301df6,

title = "HEAT: Hyperbolic Embedding of Attributed Networks",

abstract = "Finding a low dimensional representation of hierarchical, structured data described by a network remains a challenging problem in the machine learning community. An emerging approach is embedding networks into hyperbolic space since it can naturally represent a network{\textquoteright}s hierarchical structure. However, existing hyperbolic embedding approaches cannot deal with attributed networks, where nodes are annotated with additional attributes. To overcome this, we propose HEAT (for Hyperbolic Embedding of Attributed Networks). HEAT first extracts training samples from the network that captures both topological and attribute node similarity and then learns a low-dimensional hyperboloid embedding using full Riemannian Stochastic Gradient Descent. We show that HEAT can outperform other network embedding algorithms on several common downstream tasks. As a general network embedding method, HEAT opens the door to hyperbolic manifold learning on a wide range of both attributed and unattributed networks.",

keywords = "Network embedding, Hyperbolic embedding, Random walk",

author = "David McDonald and Shan He",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; 21th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2020 ; Conference date: 04-11-2020 Through 06-11-2020",

year = "2020",

month = oct,

day = "30",

doi = "10.1007/978-3-030-62362-3_4",

language = "English",

isbn = "9783030623616",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "28--40",

editor = "Cesar Analide and Paulo Novais and David Camacho and Hujun Yin",

booktitle = "Intelligent Data Engineering and Automated Learning – IDEAL 2020",

edition = "1",

}

McDonald, D & He, S 2020, HEAT: Hyperbolic Embedding of Attributed Networks. in C Analide, P Novais, D Camacho & H Yin (eds), Intelligent Data Engineering and Automated Learning – IDEAL 2020: 21st International Conference, Guimaraes, Portugal, November 4–6, 2020, Proceedings, Part I. 1 edn, Lecture Notes in Computer Science, vol. 12489, Springer, pp. 28-40, 21th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2020, Guimaraes, Portugal, 4/11/20. https://doi.org/10.1007/978-3-030-62362-3_4

HEAT: Hyperbolic Embedding of Attributed Networks. / McDonald, David; He, Shan.
Intelligent Data Engineering and Automated Learning – IDEAL 2020: 21st International Conference, Guimaraes, Portugal, November 4–6, 2020, Proceedings, Part I. ed. / Cesar Analide; Paulo Novais; David Camacho; Hujun Yin. 1. ed. Springer, 2020. p. 28-40 (Lecture Notes in Computer Science; Vol. 12489).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - HEAT

T2 - 21th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2020

AU - McDonald, David

AU - He, Shan

PY - 2020/10/30

Y1 - 2020/10/30

N2 - Finding a low dimensional representation of hierarchical, structured data described by a network remains a challenging problem in the machine learning community. An emerging approach is embedding networks into hyperbolic space since it can naturally represent a network’s hierarchical structure. However, existing hyperbolic embedding approaches cannot deal with attributed networks, where nodes are annotated with additional attributes. To overcome this, we propose HEAT (for Hyperbolic Embedding of Attributed Networks). HEAT first extracts training samples from the network that captures both topological and attribute node similarity and then learns a low-dimensional hyperboloid embedding using full Riemannian Stochastic Gradient Descent. We show that HEAT can outperform other network embedding algorithms on several common downstream tasks. As a general network embedding method, HEAT opens the door to hyperbolic manifold learning on a wide range of both attributed and unattributed networks.

AB - Finding a low dimensional representation of hierarchical, structured data described by a network remains a challenging problem in the machine learning community. An emerging approach is embedding networks into hyperbolic space since it can naturally represent a network’s hierarchical structure. However, existing hyperbolic embedding approaches cannot deal with attributed networks, where nodes are annotated with additional attributes. To overcome this, we propose HEAT (for Hyperbolic Embedding of Attributed Networks). HEAT first extracts training samples from the network that captures both topological and attribute node similarity and then learns a low-dimensional hyperboloid embedding using full Riemannian Stochastic Gradient Descent. We show that HEAT can outperform other network embedding algorithms on several common downstream tasks. As a general network embedding method, HEAT opens the door to hyperbolic manifold learning on a wide range of both attributed and unattributed networks.

KW - Network embedding

KW - Hyperbolic embedding

KW - Random walk

UR - http://www.scopus.com/inward/record.url?scp=85097442941&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-62362-3_4

DO - 10.1007/978-3-030-62362-3_4

M3 - Conference contribution

AN - SCOPUS:85097442941

SN - 9783030623616

T3 - Lecture Notes in Computer Science

SP - 28

EP - 40

BT - Intelligent Data Engineering and Automated Learning – IDEAL 2020

A2 - Analide, Cesar

A2 - Novais, Paulo

A2 - Camacho, David

A2 - Yin, Hujun

PB - Springer

Y2 - 4 November 2020 through 6 November 2020

ER -

McDonald D, He S. HEAT: Hyperbolic Embedding of Attributed Networks. In Analide C, Novais P, Camacho D, Yin H, editors, Intelligent Data Engineering and Automated Learning – IDEAL 2020: 21st International Conference, Guimaraes, Portugal, November 4–6, 2020, Proceedings, Part I. 1 ed. Springer. 2020. p. 28-40. (Lecture Notes in Computer Science). Epub 2020 Oct 27. doi: 10.1007/978-3-030-62362-3_4

HEAT: Hyperbolic Embedding of Attributed Networks

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this