Simple Rule Injection for ComplEx Embeddings

Simple Rule Injection for ComplEx Embeddings

Recent works in neural knowledge graph inference attempt to combine logic rules with knowledge graph embeddings to benefit from prior knowledge. However, they usually cannot avoid rule grounding, and injecting a diverse set of rules has still not been thoroughly explored. In this work, we propose In...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ma, Haodi, Colas, Anthony, Wang, Yuejie, Sadeghian, Ali, Wang, Daisy Zhe
Format: Artikel
Sprache:eng
Schlagworte:
Computer Science - Artificial Intelligence
Computer Science - Computation and Language
Computer Science - Learning
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Ma, Haodi
Colas, Anthony
Wang, Yuejie
Sadeghian, Ali
Wang, Daisy Zhe
description Recent works in neural knowledge graph inference attempt to combine logic rules with knowledge graph embeddings to benefit from prior knowledge. However, they usually cannot avoid rule grounding, and injecting a diverse set of rules has still not been thoroughly explored. In this work, we propose InjEx, a mechanism to inject multiple types of rules through simple constraints, which capture definite Horn rules. To start, we theoretically prove that InjEx can inject such rules. Next, to demonstrate that InjEx infuses interpretable prior knowledge into the embedding space, we evaluate InjEx on both the knowledge graph completion (KGC) and few-shot knowledge graph completion (FKGC) settings. Our experimental results reveal that InjEx outperforms both baseline KGC models as well as specialized few-shot models while maintaining its scalability and efficiency.
doi_str_mv 10.48550/arxiv.2308.03269
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2308_03269</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2308_03269</sourcerecordid><originalsourceid>FETCH-LOGICAL-a679-befc6f417ae9cfcaa326c6cad6c0bcfd0a858791872aed28ffc7994edb3b1c813</originalsourceid><addsrcrecordid>eNotjstuwjAURL1hUUE_oCvyASTYceLHEkWBIiEhAfvo-tpGRnkp0Ir-PSmwmVmMdOYQ8sVokqk8p0sY7uE3STlVCeWp0B9kcQxNX7vo8DPGtr04vIWujXw3REU3LuU9KhvjrA3t-TojEw_11X2-e0pO6_JUfMe7_WZbrHYxCKlj4zwKnzEJTqNHgPEJBYIVSA16S0HlSmqmZArOpsp7lFpnzhpuGCrGp2T-wj51q34IDQx_1b929dTmD0itPaw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Simple Rule Injection for ComplEx Embeddings</title><source>arXiv.org</source><creator>Ma, Haodi ; Colas, Anthony ; Wang, Yuejie ; Sadeghian, Ali ; Wang, Daisy Zhe</creator><creatorcontrib>Ma, Haodi ; Colas, Anthony ; Wang, Yuejie ; Sadeghian, Ali ; Wang, Daisy Zhe</creatorcontrib><description>Recent works in neural knowledge graph inference attempt to combine logic rules with knowledge graph embeddings to benefit from prior knowledge. However, they usually cannot avoid rule grounding, and injecting a diverse set of rules has still not been thoroughly explored. In this work, we propose InjEx, a mechanism to inject multiple types of rules through simple constraints, which capture definite Horn rules. To start, we theoretically prove that InjEx can inject such rules. Next, to demonstrate that InjEx infuses interpretable prior knowledge into the embedding space, we evaluate InjEx on both the knowledge graph completion (KGC) and few-shot knowledge graph completion (FKGC) settings. Our experimental results reveal that InjEx outperforms both baseline KGC models as well as specialized few-shot models while maintaining its scalability and efficiency.</description><identifier>DOI: 10.48550/arxiv.2308.03269</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computation and Language ; Computer Science - Learning</subject><creationdate>2023-08</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2308.03269$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2308.03269$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Ma, Haodi</creatorcontrib><creatorcontrib>Colas, Anthony</creatorcontrib><creatorcontrib>Wang, Yuejie</creatorcontrib><creatorcontrib>Sadeghian, Ali</creatorcontrib><creatorcontrib>Wang, Daisy Zhe</creatorcontrib><title>Simple Rule Injection for ComplEx Embeddings</title><description>Recent works in neural knowledge graph inference attempt to combine logic rules with knowledge graph embeddings to benefit from prior knowledge. However, they usually cannot avoid rule grounding, and injecting a diverse set of rules has still not been thoroughly explored. In this work, we propose InjEx, a mechanism to inject multiple types of rules through simple constraints, which capture definite Horn rules. To start, we theoretically prove that InjEx can inject such rules. Next, to demonstrate that InjEx infuses interpretable prior knowledge into the embedding space, we evaluate InjEx on both the knowledge graph completion (KGC) and few-shot knowledge graph completion (FKGC) settings. Our experimental results reveal that InjEx outperforms both baseline KGC models as well as specialized few-shot models while maintaining its scalability and efficiency.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotjstuwjAURL1hUUE_oCvyASTYceLHEkWBIiEhAfvo-tpGRnkp0Ir-PSmwmVmMdOYQ8sVokqk8p0sY7uE3STlVCeWp0B9kcQxNX7vo8DPGtr04vIWujXw3REU3LuU9KhvjrA3t-TojEw_11X2-e0pO6_JUfMe7_WZbrHYxCKlj4zwKnzEJTqNHgPEJBYIVSA16S0HlSmqmZArOpsp7lFpnzhpuGCrGp2T-wj51q34IDQx_1b929dTmD0itPaw</recordid><startdate>20230806</startdate><enddate>20230806</enddate><creator>Ma, Haodi</creator><creator>Colas, Anthony</creator><creator>Wang, Yuejie</creator><creator>Sadeghian, Ali</creator><creator>Wang, Daisy Zhe</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230806</creationdate><title>Simple Rule Injection for ComplEx Embeddings</title><author>Ma, Haodi ; Colas, Anthony ; Wang, Yuejie ; Sadeghian, Ali ; Wang, Daisy Zhe</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a679-befc6f417ae9cfcaa326c6cad6c0bcfd0a858791872aed28ffc7994edb3b1c813</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Ma, Haodi</creatorcontrib><creatorcontrib>Colas, Anthony</creatorcontrib><creatorcontrib>Wang, Yuejie</creatorcontrib><creatorcontrib>Sadeghian, Ali</creatorcontrib><creatorcontrib>Wang, Daisy Zhe</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Ma, Haodi</au><au>Colas, Anthony</au><au>Wang, Yuejie</au><au>Sadeghian, Ali</au><au>Wang, Daisy Zhe</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Simple Rule Injection for ComplEx Embeddings</atitle><date>2023-08-06</date><risdate>2023</risdate><abstract>Recent works in neural knowledge graph inference attempt to combine logic rules with knowledge graph embeddings to benefit from prior knowledge. However, they usually cannot avoid rule grounding, and injecting a diverse set of rules has still not been thoroughly explored. In this work, we propose InjEx, a mechanism to inject multiple types of rules through simple constraints, which capture definite Horn rules. To start, we theoretically prove that InjEx can inject such rules. Next, to demonstrate that InjEx infuses interpretable prior knowledge into the embedding space, we evaluate InjEx on both the knowledge graph completion (KGC) and few-shot knowledge graph completion (FKGC) settings. Our experimental results reveal that InjEx outperforms both baseline KGC models as well as specialized few-shot models while maintaining its scalability and efficiency.</abstract><doi>10.48550/arxiv.2308.03269</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2308.03269
ispartof
issn
language eng
recordid cdi_arxiv_primary_2308_03269
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Computation and Language
Computer Science - Learning
title Simple Rule Injection for ComplEx Embeddings
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-09T05%3A28%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Simple%20Rule%20Injection%20for%20ComplEx%20Embeddings&rft.au=Ma,%20Haodi&rft.date=2023-08-06&rft_id=info:doi/10.48550/arxiv.2308.03269&rft_dat=%3Carxiv_GOX%3E2308_03269%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true