Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206313911> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3206313911 abstract "A knowledge graph (KG) is a data structure which represents entities and relations as the vertices and edges of a directed graph with edge types. KGs are an important primitive in modern machine learning and artificial intelligence. Embedding-based models, such as the seminal TransE [Bordes et al., 2013] and the recent PairRE [Chao et al., 2020] are among the most popular and successful approaches for representing KGs and inferring missing edges (link completion). Their relative success is often credited in the literature to their ability to learn logical rules between the relations. In this work, we investigate whether learning rules between relations is indeed what drives the performance of embedding-based methods. We define motif learning and two alternative mechanisms, network learning (based only on the connectivity of the KG, ignoring the relation types), and unstructured statistical learning (ignoring the connectivity of the graph). Using experiments on synthetic KGs, we show that KG models can learn motifs and how this ability is degraded by non-motif (noise) edges. We propose tests to distinguish the contributions of the three mechanisms to performance, and apply them to popular KG benchmarks. We also discuss an issue with the standard performance testing protocol and suggest an improvement. To appear in the proceedings of Complex Networks 2021." @default.
- W3206313911 created "2021-10-25" @default.
- W3206313911 creator A5040335961 @default.
- W3206313911 creator A5044142637 @default.
- W3206313911 creator A5046098711 @default.
- W3206313911 creator A5053210820 @default.
- W3206313911 creator A5056770872 @default.
- W3206313911 creator A5065085371 @default.
- W3206313911 creator A5083523912 @default.
- W3206313911 date "2021-10-19" @default.
- W3206313911 modified "2023-09-27" @default.
- W3206313911 title "What is Learned in Knowledge Graph Embeddings" @default.
- W3206313911 cites W1523004878 @default.
- W3206313911 cites W1552847225 @default.
- W3206313911 cites W1663973292 @default.
- W3206313911 cites W2117446654 @default.
- W3206313911 cites W2123228027 @default.
- W3206313911 cites W2127795553 @default.
- W3206313911 cites W2166602562 @default.
- W3206313911 cites W2798955995 @default.
- W3206313911 cites W2889583850 @default.
- W3206313911 cites W2949434543 @default.
- W3206313911 cites W2951077644 @default.
- W3206313911 cites W2963119656 @default.
- W3206313911 cites W2963432357 @default.
- W3206313911 cites W3005977214 @default.
- W3206313911 cites W3023207061 @default.
- W3206313911 cites W3028709955 @default.
- W3206313911 cites W3032390337 @default.
- W3206313911 cites W3034374701 @default.
- W3206313911 cites W3040868708 @default.
- W3206313911 cites W3045272257 @default.
- W3206313911 cites W3103296573 @default.
- W3206313911 cites W3122822969 @default.
- W3206313911 cites W3175989614 @default.
- W3206313911 cites W3021975806 @default.
- W3206313911 hasPublicationYear "2021" @default.
- W3206313911 type Work @default.
- W3206313911 sameAs 3206313911 @default.
- W3206313911 citedByCount "0" @default.
- W3206313911 crossrefType "posted-content" @default.
- W3206313911 hasAuthorship W3206313911A5040335961 @default.
- W3206313911 hasAuthorship W3206313911A5044142637 @default.
- W3206313911 hasAuthorship W3206313911A5046098711 @default.
- W3206313911 hasAuthorship W3206313911A5053210820 @default.
- W3206313911 hasAuthorship W3206313911A5056770872 @default.
- W3206313911 hasAuthorship W3206313911A5065085371 @default.
- W3206313911 hasAuthorship W3206313911A5083523912 @default.
- W3206313911 hasConcept C119857082 @default.
- W3206313911 hasConcept C121332964 @default.
- W3206313911 hasConcept C132525143 @default.
- W3206313911 hasConcept C154945302 @default.
- W3206313911 hasConcept C24890656 @default.
- W3206313911 hasConcept C2987255567 @default.
- W3206313911 hasConcept C32276052 @default.
- W3206313911 hasConcept C41008148 @default.
- W3206313911 hasConcept C41608201 @default.
- W3206313911 hasConcept C80444323 @default.
- W3206313911 hasConceptScore W3206313911C119857082 @default.
- W3206313911 hasConceptScore W3206313911C121332964 @default.
- W3206313911 hasConceptScore W3206313911C132525143 @default.
- W3206313911 hasConceptScore W3206313911C154945302 @default.
- W3206313911 hasConceptScore W3206313911C24890656 @default.
- W3206313911 hasConceptScore W3206313911C2987255567 @default.
- W3206313911 hasConceptScore W3206313911C32276052 @default.
- W3206313911 hasConceptScore W3206313911C41008148 @default.
- W3206313911 hasConceptScore W3206313911C41608201 @default.
- W3206313911 hasConceptScore W3206313911C80444323 @default.
- W3206313911 hasLocation W32063139111 @default.
- W3206313911 hasOpenAccess W3206313911 @default.
- W3206313911 hasPrimaryLocation W32063139111 @default.
- W3206313911 hasRelatedWork W2508478068 @default.
- W3206313911 hasRelatedWork W2767688584 @default.
- W3206313911 hasRelatedWork W2784887575 @default.
- W3206313911 hasRelatedWork W2949186007 @default.
- W3206313911 hasRelatedWork W2962869320 @default.
- W3206313911 hasRelatedWork W2963169753 @default.
- W3206313911 hasRelatedWork W2997262687 @default.
- W3206313911 hasRelatedWork W2997932512 @default.
- W3206313911 hasRelatedWork W3005588274 @default.
- W3206313911 hasRelatedWork W3017152929 @default.
- W3206313911 hasRelatedWork W3035021589 @default.
- W3206313911 hasRelatedWork W3047602953 @default.
- W3206313911 hasRelatedWork W3088093011 @default.
- W3206313911 hasRelatedWork W3088159601 @default.
- W3206313911 hasRelatedWork W3104211877 @default.
- W3206313911 hasRelatedWork W3126412650 @default.
- W3206313911 hasRelatedWork W3129756592 @default.
- W3206313911 hasRelatedWork W3145277743 @default.
- W3206313911 hasRelatedWork W3146861612 @default.
- W3206313911 hasRelatedWork W3210658351 @default.
- W3206313911 isParatext "false" @default.
- W3206313911 isRetracted "false" @default.
- W3206313911 magId "3206313911" @default.
- W3206313911 workType "article" @default.