Research map / Graph learning / Knowledge and temporal graphs
Multi-relational Poincaré 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 Poincaré embeddings instead of Euclidean embeddings | MuRE/MuRP · WN18RR | MRR: .475 -> .481 | Helped |
| use Poincaré embeddings instead of Euclidean embeddings | MuRE/MuRP · FB15k-237 | MRR: .336 -> .335 | Hurt |
| use Poincaré embeddings instead of Euclidean embeddings | MuRE/MuRP · NELL-995-h100 d=40 and d=200 | MRR: .330 -> .344; .355 -> .360 | Helped |
| use Poincaré embeddings instead of Euclidean embeddings | MuRE/MuRP · NELL-995-h25 d=200 | MRR: .365 -> .359 | Hurt |
| remove subject relation matrix (ablation) | MuRP · WN18RR | MRR: .477 -> .307 | Hurt |
| remove object relation translation (ablation) | MuRP · WN18RR | MRR: .477 -> .401 | Hurt |
| use separate subject and object relation matrices (ablation) | MuRP · WN18RR | MRR: .477 -> .367 | Hurt |
| translate subject and transform object instead (ablation) | MuRP · WN18RR | MRR: .477 -> .454 | Hurt |
| remove object bias (ablation) | MuRP · WN18RR | MRR: .477 -> .463 | Hurt |
| remove subject bias (ablation) | MuRP · WN18RR | MRR: .477 -> .460 | Hurt |
| replace biases with squared entity embedding norms (ablation) | MuRP · WN18RR | MRR: .477 -> .414 | Hurt |
| replace biases with squared transformed embedding norms (ablation) | MuRP · WN18RR | MRR: .477 -> .434 | Hurt |
Closest papers by meaning
- Poincaré Embeddings for Learning Hierarchical Representations (NeurIPS 2017 · similarity 0.88)
- Geometry Interaction Knowledge Graph Embeddings (AAAI 2022 · similarity 0.88)
- Learning Hierarchy-Aware Knowledge Graph Embeddings for Link Prediction (AAAI 2020 · similarity 0.87)
- Weighted Embeddings for Low-Dimensional Graph Representation (AAAI 2025 · similarity 0.87)
- Hyperbolic Embeddings of Supervised Models (NeurIPS 2024 · similarity 0.86)
- Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization (ICML 2024 · similarity 0.86)
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.