Research map / Graph learning / Knowledge and temporal graphs
Quaternion Knowledge Graph Embeddings
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use quaternion embeddings and relational rotation instead of complex-space embeddings | QuatE · WN18RR | MRR: 0.476 -> 0.488 | Helped |
| Use N3 regularization and reciprocal learning (ablation) | QuatE · FB15K-237 | MRR: 0.311 -> 0.366 | Helped |
| Remove relation-quaternion normalization (ablation) | QuatE · WN18 | MRR: 0.950 -> 0.936 | Hurt |
| Replace relational rotation with W_r · (Q_h ⊗ Q_t) (ablation) | QuatE · WN18 | MRR: 0.950 -> 0.784 | Hurt |
| Add a relation rotation to the tail entity (ablation) | QuatE · WN18RR | MRR: 0.488 -> 0.477 | Hurt |
| Extend the embedding algebra from quaternions to octonions (ablation) | QuatE · WN18 | MRR: 0.950 -> 0.950 | No clear effect |
| Use type constraints (ablation) | QuatE · FB15K-237 | MRR: 0.311 -> 0.348 | Helped |
Closest papers by meaning
- Dual Quaternion Knowledge Graph Embeddings (AAAI 2021 · similarity 0.94)
- RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space (ICLR 2019 · similarity 0.87)
- Geometry Interaction Knowledge Graph Embeddings (AAAI 2022 · similarity 0.87)
- Improved Knowledge Graph Embedding Using Background Taxonomic Information (AAAI 2019 · similarity 0.85)
- NestE: Modeling Nested Relational Structures for Knowledge Graph Reasoning (AAAI 2024 · similarity 0.84)
- Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization (ICML 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.