Efficient parallel translating embedding for knowledge graphs

Denghui Zhang, Manling Li, Yantao Jia, Yuanzhuo Wang, Xueqi Cheng

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

14 Scopus citations

Abstract

Knowledge graph embedding aims to embed entities and relations of knowledge graphs into low-dimensional vector spaces. Translating embedding methods regard relations as the translation from head entities to tail entities, which achieve the state-of-The-Art results among knowledge graph embedding methods. However, a major limitation of these methods is the time consuming training process, which may take several days or even weeks for large knowledge graphs, and result in great difficulty in practical applications. In this paper, we propose an effcient parallel framework for translating embedding methods, called ParTrans-X, which enables the methods to be paralleled without locks by utilizing the distinguished structures of knowledge graphs. Experiments on two datasets with three typical translating embedding methods, i.e., TransE [3], TransH [19], and a more efficient variant TransE- AdaGrad [11] validate that ParTrans-X can speed up the training process by more than an order of magnitude.

Original languageEnglish
Title of host publicationProceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017
Pages460-468
Number of pages9
ISBN (Electronic)9781450349512
DOIs
StatePublished - 23 Aug 2017
Event16th IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017 - Leipzig, Germany
Duration: 23 Aug 201726 Aug 2017

Publication series

NameProceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017

Conference

Conference16th IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017
Country/TerritoryGermany
CityLeipzig
Period23/08/1726/08/17

Keywords

  • Knowledge graph embedding
  • Parallel
  • Translation-based

Fingerprint

Dive into the research topics of 'Efficient parallel translating embedding for knowledge graphs'. Together they form a unique fingerprint.

Cite this