A Recurrent Graph Neural Network for Multi-relational Data

Vassilis N. Ioannidis, Antonio G. Marques, Georgios B Giannakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

The era of data deluge has sparked the interest in graph-based learning methods in a number of disciplines such as sociology, biology, neuroscience, or engineering. In this paper, we introduce a graph recurrent neural network (GRNN) for scalable semi-supervised learning from multi-relational data. Key aspects of the novel GRNN architecture are the use of multi-relational graphs, the dynamic adaptation to the different relations via learnable weights, and the consideration of graph-based regularizers to promote smoothness and alleviate over-parametrization. Our ultimate goal is to design a powerful learning architecture able to: discover complex and highly non-linear data associations, combine (and select) multiple types of relations, and scale gracefully with respect to the size of the graph. Numerical tests with real datasets corroborate the design goals and illustrate the performance gains relative to competing alternatives.

Original languageEnglish (US)
Title of host publication2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages8157-8161
Number of pages5
ISBN (Electronic)9781479981311
DOIs
StatePublished - May 2019
Event44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Brighton, United Kingdom
Duration: May 12 2019May 17 2019

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2019-May
ISSN (Print)1520-6149

Conference

Conference44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
CountryUnited Kingdom
CityBrighton
Period5/12/195/17/19

Fingerprint

Recurrent neural networks
Neural networks
Supervised learning
Network architecture

Keywords

  • Deep neural networks
  • graph recurrent neural networks
  • graph signals
  • multi-relational graphs

Cite this

Ioannidis, V. N., Marques, A. G., & Giannakis, G. B. (2019). A Recurrent Graph Neural Network for Multi-relational Data. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings (pp. 8157-8161). [8682836] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2019.8682836

A Recurrent Graph Neural Network for Multi-relational Data. / Ioannidis, Vassilis N.; Marques, Antonio G.; Giannakis, Georgios B.

2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 8157-8161 8682836 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ioannidis, VN, Marques, AG & Giannakis, GB 2019, A Recurrent Graph Neural Network for Multi-relational Data. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings., 8682836, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2019-May, Institute of Electrical and Electronics Engineers Inc., pp. 8157-8161, 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019, Brighton, United Kingdom, 5/12/19. https://doi.org/10.1109/ICASSP.2019.8682836
Ioannidis VN, Marques AG, Giannakis GB. A Recurrent Graph Neural Network for Multi-relational Data. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 8157-8161. 8682836. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2019.8682836
Ioannidis, Vassilis N. ; Marques, Antonio G. ; Giannakis, Georgios B. / A Recurrent Graph Neural Network for Multi-relational Data. 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 8157-8161 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{2d0612c4feaa485a94a88140379cb33e,
title = "A Recurrent Graph Neural Network for Multi-relational Data",
abstract = "The era of data deluge has sparked the interest in graph-based learning methods in a number of disciplines such as sociology, biology, neuroscience, or engineering. In this paper, we introduce a graph recurrent neural network (GRNN) for scalable semi-supervised learning from multi-relational data. Key aspects of the novel GRNN architecture are the use of multi-relational graphs, the dynamic adaptation to the different relations via learnable weights, and the consideration of graph-based regularizers to promote smoothness and alleviate over-parametrization. Our ultimate goal is to design a powerful learning architecture able to: discover complex and highly non-linear data associations, combine (and select) multiple types of relations, and scale gracefully with respect to the size of the graph. Numerical tests with real datasets corroborate the design goals and illustrate the performance gains relative to competing alternatives.",
keywords = "Deep neural networks, graph recurrent neural networks, graph signals, multi-relational graphs",
author = "Ioannidis, {Vassilis N.} and Marques, {Antonio G.} and Giannakis, {Georgios B}",
year = "2019",
month = "5",
doi = "10.1109/ICASSP.2019.8682836",
language = "English (US)",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "8157--8161",
booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings",

}

TY - GEN

T1 - A Recurrent Graph Neural Network for Multi-relational Data

AU - Ioannidis, Vassilis N.

AU - Marques, Antonio G.

AU - Giannakis, Georgios B

PY - 2019/5

Y1 - 2019/5

N2 - The era of data deluge has sparked the interest in graph-based learning methods in a number of disciplines such as sociology, biology, neuroscience, or engineering. In this paper, we introduce a graph recurrent neural network (GRNN) for scalable semi-supervised learning from multi-relational data. Key aspects of the novel GRNN architecture are the use of multi-relational graphs, the dynamic adaptation to the different relations via learnable weights, and the consideration of graph-based regularizers to promote smoothness and alleviate over-parametrization. Our ultimate goal is to design a powerful learning architecture able to: discover complex and highly non-linear data associations, combine (and select) multiple types of relations, and scale gracefully with respect to the size of the graph. Numerical tests with real datasets corroborate the design goals and illustrate the performance gains relative to competing alternatives.

AB - The era of data deluge has sparked the interest in graph-based learning methods in a number of disciplines such as sociology, biology, neuroscience, or engineering. In this paper, we introduce a graph recurrent neural network (GRNN) for scalable semi-supervised learning from multi-relational data. Key aspects of the novel GRNN architecture are the use of multi-relational graphs, the dynamic adaptation to the different relations via learnable weights, and the consideration of graph-based regularizers to promote smoothness and alleviate over-parametrization. Our ultimate goal is to design a powerful learning architecture able to: discover complex and highly non-linear data associations, combine (and select) multiple types of relations, and scale gracefully with respect to the size of the graph. Numerical tests with real datasets corroborate the design goals and illustrate the performance gains relative to competing alternatives.

KW - Deep neural networks

KW - graph recurrent neural networks

KW - graph signals

KW - multi-relational graphs

UR - http://www.scopus.com/inward/record.url?scp=85068973407&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068973407&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2019.8682836

DO - 10.1109/ICASSP.2019.8682836

M3 - Conference contribution

AN - SCOPUS:85068973407

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 8157

EP - 8161

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -