Papers › Quaternion Knowledge Graph Embeddings

Quaternion Knowledge Graph Embeddings

23 Apr 2019NeurIPS 2019 12arXiv:1904.10281archive 2025-07-28

Shuai Zhang, Yi Tay, Lina Yao, Qi Liu

In this work, we move beyond the traditional complex-valued representations, introducing more expressive hypercomplex representations to model entities and relations for knowledge graph embeddings. More specifically, quaternion embeddings, hypercomplex-valued embeddings with three imaginary components, are utilized to represent entities. Relations are modelled as rotations in the quaternion space. The advantages of the proposed approach are: (1) Latent inter-dependencies (between all components) are aptly captured with Hamilton product, encouraging a more compact interaction between entities and relations; (2) Quaternions enable expressive rotation in four-dimensional space and have more degree of freedom than rotation in complex plane; (3) The proposed framework is a generalization of ComplEx on hypercomplex space while offering better geometrical interpretations, concurrently satisfying the key desiderata of relational representation learning (i.e., modeling symmetry, anti-symmetry and inversion). Experimental results demonstrate that our method achieves state-of-the-art performance on four well-established knowledge graph completion benchmarks.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

cheungdaven/QuatE mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Graph EmbeddingKnowledge Graph CompletionKnowledge Graph EmbeddingKnowledge Graph EmbeddingsKnowledge GraphsLink PredictionRepresentation Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Link Prediction FB15k QuatE Hits@1 0.800 #3 of 23 Archive leaderboard report
Link Prediction FB15k QuatE Hits@10 0.900 #3 of 23 Archive leaderboard report
Link Prediction FB15k QuatE Hits@3 0.859 #3 of 23 Archive leaderboard report
Link Prediction FB15k QuatE MR 17 #3 of 23 Archive leaderboard report
Link Prediction FB15k QuatE MRR 0.833 #3 of 23 Archive leaderboard report
Link Prediction FB15k-237 QuatE Hits@1 0.248 #38 of 75 Archive leaderboard report
Link Prediction FB15k-237 QuatE Hits@10 0.550 #38 of 75 Archive leaderboard report
Link Prediction FB15k-237 QuatE Hits@3 0.382 #38 of 75 Archive leaderboard report
Link Prediction FB15k-237 QuatE MR 87 #38 of 75 Archive leaderboard report
Link Prediction FB15k-237 QuatE MRR 0.348 #38 of 75 Archive leaderboard report
Link Prediction WN18 QuatE Hits@1 0.945 #7 of 37 Archive leaderboard report
Link Prediction WN18 QuatE Hits@10 0.959 #7 of 37 Archive leaderboard report
Link Prediction WN18 QuatE Hits@3 0.954 #7 of 37 Archive leaderboard report
Link Prediction WN18 QuatE MR 162 #7 of 37 Archive leaderboard report
Link Prediction WN18 QuatE MRR 0.95 #7 of 37 Archive leaderboard report
Link Prediction WN18RR QuatE Hits@1 0.438 #24 of 75 Archive leaderboard report
Link Prediction WN18RR QuatE Hits@10 0.582 #24 of 75 Archive leaderboard report
Link Prediction WN18RR QuatE Hits@3 0.508 #24 of 75 Archive leaderboard report
Link Prediction WN18RR QuatE MR 2314 #24 of 75 Archive leaderboard report
Link Prediction WN18RR QuatE MRR 0.488 #24 of 75 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections