Graph Attention Auto-Encoders

Amin Salehi, Hasan Davulcu

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

32 Scopus citations

Abstract

Auto-encoders have emerged as a successful framework for unsupervised learning. However, conventional auto-encoders are incapable of utilizing explicit relations in structured data. To take advantage of relations in graph-structured data, several graph auto-encoders have recently been proposed, but they neglect to reconstruct either the graph structure or node attributes. In this paper, we present the graph attention auto-encoder (GATE), a neural network architecture for unsupervised representation learning on graph-structured data. Our architecture is able to reconstruct graph-structured inputs, including both node attributes and the graph structure, through stacked encoder/decoder layers equipped with self-Attention mechanisms. In the encoder, by considering node attributes as initial node representations, each layer generates new representations of nodes by attending over their neighbors' representations. In the decoder, we attempt to reverse the encoding process to reconstruct node attributes. Moreover, node representations are regularized to reconstruct the graph structure. Our proposed architecture does not need to know the graph structure upfront, and thus it can be applied to inductive learning. Our experiments demonstrate competitive performance on several node classification benchmark datasets for transductive and inductive tasks, even exceeding the performance of supervised learning baselines in most cases.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE 32nd International Conference on Tools with Artificial Intelligence, ICTAI 2020
EditorsMiltos Alamaniotis, Shimei Pan
PublisherIEEE Computer Society
Pages989-996
Number of pages8
ISBN (Electronic)9781728192284
DOIs
StatePublished - Nov 2020
Event32nd IEEE International Conference on Tools with Artificial Intelligence, ICTAI 2020 - Virtual, Baltimore, United States
Duration: Nov 9 2020Nov 11 2020

Publication series

NameProceedings - International Conference on Tools with Artificial Intelligence, ICTAI
Volume2020-November
ISSN (Print)1082-3409

Conference

Conference32nd IEEE International Conference on Tools with Artificial Intelligence, ICTAI 2020
Country/TerritoryUnited States
CityVirtual, Baltimore
Period11/9/2011/11/20

Keywords

  • n/a

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Graph Attention Auto-Encoders'. Together they form a unique fingerprint.

Cite this