{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/efficient-parallel-translating-embedding-for","title":"Efficient Parallel Translating Embedding For Knowledge Graphs","arxiv_id":"1703.10316","date":"2017-03-30","proceeding":null,"authors":["Denghui Zhang","Manling Li","Yantao Jia","Yuanzhuo Wang","Xue-Qi Cheng"],"abstract":"Knowledge graph embedding aims to embed entities and relations of knowledge\ngraphs into low-dimensional vector spaces. Translating embedding methods regard\nrelations as the translation from head entities to tail entities, which achieve\nthe state-of-the-art results among knowledge graph embedding methods. However,\na major limitation of these methods is the time consuming training process,\nwhich may take several days or even weeks for large knowledge graphs, and\nresult in great difficulty in practical applications. In this paper, we propose\nan efficient parallel framework for translating embedding methods, called\nParTrans-X, which enables the methods to be paralleled without locks by\nutilizing the distinguished structures of knowledge graphs. Experiments on two\ndatasets with three typical translating embedding methods, i.e., TransE [3],\nTransH [17], and a more efficient variant TransE- AdaGrad [10] validate that\nParTrans-X can speed up the training process by more than an order of\nmagnitude.","url_abs":"http://arxiv.org/abs/1703.10316v4","url_pdf":"http://arxiv.org/pdf/1703.10316v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"efficient-parallel-translating-embedding-for","repo_url":"https://github.com/zdh2292390/ParTrans-X","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"graph-embedding","task_name":"Graph Embedding"},{"task_slug":"knowledge-graph-embedding","task_name":"Knowledge Graph Embedding"},{"task_slug":"knowledge-graphs","task_name":"Knowledge Graphs"},{"task_slug":"link-prediction","task_name":"Link Prediction"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"adagrad","method_name":"AdaGrad"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"transe","method_name":"TransE"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/link-prediction-on-fb15k-1","task":"Link Prediction","dataset":"FB15k","model":"ParTransH","rank_in_archive_order":10,"of":10,"metrics":{"Hits@10":"0.468","MR":"60"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-fb15k-filtered","task":"Link Prediction","dataset":"FB15k (filtered)","model":"ParTransH","rank_in_archive_order":1,"of":1,"metrics":{"Hits@10":"65.7","MR":"60"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-wn18","task":"Link Prediction","dataset":"WN18","model":"ParTransH","rank_in_archive_order":33,"of":37,"metrics":{"Hits@10":"0.668","MR":"215"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-wn18-filtered","task":"Link Prediction","dataset":"WN18 (filtered)","model":"ParTransH","rank_in_archive_order":1,"of":1,"metrics":{"Hits@10":"76.6","MR":"203"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}