Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning

Probabilistic logical rule learning has shown great strength in logical rule mining and knowledge graph completion. It learns logical rules to predict missing edges by reasoning on existing edges in the knowledge graph. However, previous efforts have largely been limited to only modeling chain-like...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Han, Chi, He, Qizheng, Yu, Charles, Du, Xinya, Tong, Hanghang, Ji, Heng
Format: Artikel
Sprache:eng
Schlagworte:
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 Han, Chi
He, Qizheng
Yu, Charles
Du, Xinya
Tong, Hanghang
Ji, Heng
description Probabilistic logical rule learning has shown great strength in logical rule mining and knowledge graph completion. It learns logical rules to predict missing edges by reasoning on existing edges in the knowledge graph. However, previous efforts have largely been limited to only modeling chain-like Horn clauses such as $R_1(x,z)\land R_2(z,y)\Rightarrow H(x,y)$. This formulation overlooks additional contextual information from neighboring sub-graphs of entity variables $x$, $y$ and $z$. Intuitively, there is a large gap here, as local sub-graphs have been found to provide important information for knowledge graph completion. Inspired by these observations, we propose Logical Entity RePresentation (LERP) to encode contextual information of entities in the knowledge graph. A LERP is designed as a vector of probabilistic logical functions on the entity's neighboring sub-graph. It is an interpretable representation while allowing for differentiable optimization. We can then incorporate LERP into probabilistic logical rule learning to learn more expressive rules. Empirical results demonstrate that with LERP, our model outperforms other rule learning methods in knowledge graph completion and is comparable or even superior to state-of-the-art black-box methods. Moreover, we find that our model can discover a more expressive family of logical rules. LERP can also be further combined with embedding learning methods like TransE to make it more interpretable.
doi_str_mv 10.48550/arxiv.2305.12738
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2305_12738</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2305_12738</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-f10008bdbb458b3bc16e80e89252dcdde437906f6e5ca323e68ac38d082761d43</originalsourceid><addsrcrecordid>eNotz8tOwzAUBFBvWKDCB7DCP5DgR-y4y6qUgoiEFHUfXdvXwVLqRE549O8phc3MZjTSIeSOs7IySrEHyN_xsxSSqZKLWppr0jZjHx0MdJeWuJxoi1PGGdMCSxwTjYm-pvFrQN9jsc8wvc80jJk-xhAwn2cR7IC0_ThHg5BTTP0NuQowzHj73ytyeNodts9F87Z_2W6aAnRtisAZY8Z6aytlrLSOazQMzVoo4Z33WMl6zXTQqBxIIVEbcNJ4ZkStua_kitz_3V5Q3ZTjEfKp-8V1F5z8ATInSpA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning</title><source>arXiv.org</source><creator>Han, Chi ; He, Qizheng ; Yu, Charles ; Du, Xinya ; Tong, Hanghang ; Ji, Heng</creator><creatorcontrib>Han, Chi ; He, Qizheng ; Yu, Charles ; Du, Xinya ; Tong, Hanghang ; Ji, Heng</creatorcontrib><description>Probabilistic logical rule learning has shown great strength in logical rule mining and knowledge graph completion. It learns logical rules to predict missing edges by reasoning on existing edges in the knowledge graph. However, previous efforts have largely been limited to only modeling chain-like Horn clauses such as $R_1(x,z)\land R_2(z,y)\Rightarrow H(x,y)$. This formulation overlooks additional contextual information from neighboring sub-graphs of entity variables $x$, $y$ and $z$. Intuitively, there is a large gap here, as local sub-graphs have been found to provide important information for knowledge graph completion. Inspired by these observations, we propose Logical Entity RePresentation (LERP) to encode contextual information of entities in the knowledge graph. A LERP is designed as a vector of probabilistic logical functions on the entity's neighboring sub-graph. It is an interpretable representation while allowing for differentiable optimization. We can then incorporate LERP into probabilistic logical rule learning to learn more expressive rules. Empirical results demonstrate that with LERP, our model outperforms other rule learning methods in knowledge graph completion and is comparable or even superior to state-of-the-art black-box methods. Moreover, we find that our model can discover a more expressive family of logical rules. LERP can also be further combined with embedding learning methods like TransE to make it more interpretable.</description><identifier>DOI: 10.48550/arxiv.2305.12738</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning ; Computer Science - Logic in Computer Science</subject><creationdate>2023-05</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2305.12738$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2305.12738$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Han, Chi</creatorcontrib><creatorcontrib>He, Qizheng</creatorcontrib><creatorcontrib>Yu, Charles</creatorcontrib><creatorcontrib>Du, Xinya</creatorcontrib><creatorcontrib>Tong, Hanghang</creatorcontrib><creatorcontrib>Ji, Heng</creatorcontrib><title>Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning</title><description>Probabilistic logical rule learning has shown great strength in logical rule mining and knowledge graph completion. It learns logical rules to predict missing edges by reasoning on existing edges in the knowledge graph. However, previous efforts have largely been limited to only modeling chain-like Horn clauses such as $R_1(x,z)\land R_2(z,y)\Rightarrow H(x,y)$. This formulation overlooks additional contextual information from neighboring sub-graphs of entity variables $x$, $y$ and $z$. Intuitively, there is a large gap here, as local sub-graphs have been found to provide important information for knowledge graph completion. Inspired by these observations, we propose Logical Entity RePresentation (LERP) to encode contextual information of entities in the knowledge graph. A LERP is designed as a vector of probabilistic logical functions on the entity's neighboring sub-graph. It is an interpretable representation while allowing for differentiable optimization. We can then incorporate LERP into probabilistic logical rule learning to learn more expressive rules. Empirical results demonstrate that with LERP, our model outperforms other rule learning methods in knowledge graph completion and is comparable or even superior to state-of-the-art black-box methods. Moreover, we find that our model can discover a more expressive family of logical rules. LERP can also be further combined with embedding learning methods like TransE to make it more interpretable.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><subject>Computer Science - Logic in Computer Science</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz8tOwzAUBFBvWKDCB7DCP5DgR-y4y6qUgoiEFHUfXdvXwVLqRE549O8phc3MZjTSIeSOs7IySrEHyN_xsxSSqZKLWppr0jZjHx0MdJeWuJxoi1PGGdMCSxwTjYm-pvFrQN9jsc8wvc80jJk-xhAwn2cR7IC0_ThHg5BTTP0NuQowzHj73ytyeNodts9F87Z_2W6aAnRtisAZY8Z6aytlrLSOazQMzVoo4Z33WMl6zXTQqBxIIVEbcNJ4ZkStua_kitz_3V5Q3ZTjEfKp-8V1F5z8ATInSpA</recordid><startdate>20230522</startdate><enddate>20230522</enddate><creator>Han, Chi</creator><creator>He, Qizheng</creator><creator>Yu, Charles</creator><creator>Du, Xinya</creator><creator>Tong, Hanghang</creator><creator>Ji, Heng</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230522</creationdate><title>Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning</title><author>Han, Chi ; He, Qizheng ; Yu, Charles ; Du, Xinya ; Tong, Hanghang ; Ji, Heng</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-f10008bdbb458b3bc16e80e89252dcdde437906f6e5ca323e68ac38d082761d43</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><topic>Computer Science - Logic in Computer Science</topic><toplevel>online_resources</toplevel><creatorcontrib>Han, Chi</creatorcontrib><creatorcontrib>He, Qizheng</creatorcontrib><creatorcontrib>Yu, Charles</creatorcontrib><creatorcontrib>Du, Xinya</creatorcontrib><creatorcontrib>Tong, Hanghang</creatorcontrib><creatorcontrib>Ji, Heng</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Han, Chi</au><au>He, Qizheng</au><au>Yu, Charles</au><au>Du, Xinya</au><au>Tong, Hanghang</au><au>Ji, Heng</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning</atitle><date>2023-05-22</date><risdate>2023</risdate><abstract>Probabilistic logical rule learning has shown great strength in logical rule mining and knowledge graph completion. It learns logical rules to predict missing edges by reasoning on existing edges in the knowledge graph. However, previous efforts have largely been limited to only modeling chain-like Horn clauses such as $R_1(x,z)\land R_2(z,y)\Rightarrow H(x,y)$. This formulation overlooks additional contextual information from neighboring sub-graphs of entity variables $x$, $y$ and $z$. Intuitively, there is a large gap here, as local sub-graphs have been found to provide important information for knowledge graph completion. Inspired by these observations, we propose Logical Entity RePresentation (LERP) to encode contextual information of entities in the knowledge graph. A LERP is designed as a vector of probabilistic logical functions on the entity's neighboring sub-graph. It is an interpretable representation while allowing for differentiable optimization. We can then incorporate LERP into probabilistic logical rule learning to learn more expressive rules. Empirical results demonstrate that with LERP, our model outperforms other rule learning methods in knowledge graph completion and is comparable or even superior to state-of-the-art black-box methods. Moreover, we find that our model can discover a more expressive family of logical rules. LERP can also be further combined with embedding learning methods like TransE to make it more interpretable.</abstract><doi>10.48550/arxiv.2305.12738</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2305.12738
ispartof
issn
language eng
recordid cdi_arxiv_primary_2305_12738
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Learning
Computer Science - Logic in Computer Science
title Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-24T09%3A29%3A42IST&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=Logical%20Entity%20Representation%20in%20Knowledge-Graphs%20for%20Differentiable%20Rule%20Learning&rft.au=Han,%20Chi&rft.date=2023-05-22&rft_id=info:doi/10.48550/arxiv.2305.12738&rft_dat=%3Carxiv_GOX%3E2305_12738%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