Research map / Graph learning / Knowledge and temporal graphs
LogicENN: A Neural Based Knowledge Graphs Embedding Model With Logical Rules
Mojtaba Nayyeri, Chengjin Xu, Mirza Mohtashim Alam, Jens Lehmann, Hamed Shariat Yazdi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add inverse rule injection to LogicENN_R (ablation) | LogicENN_R · FB15k (raw) | Hits@10: 42.2 -> 62.1 | Improved |
| add inverse rule injection to LogicENN_R (ablation) | LogicENN_R · FB15k (raw) | MRR: 18.1 -> 37.7 | Improved |
| add inverse rule injection to LogicENN_S (ablation) | LogicENN_S (sigmoid) · FB15k (raw) | Hits@10: 36.6 -> 59.3 | Improved |
| add implication rule (grounding-free) to LogicENN_R (ablation) | LogicENN_R · FB15k (raw) | Hits@10: 42.2 -> 40.5 | No clear effect |
| add symmetry rule to LogicENN_R (ablation) | LogicENN_R · FB15k (raw) | MR: 320 -> 299 | Improved |
| add equivalence rule to LogicENN_R (ablation) | LogicENN_R · FB15k (raw) | MR: 320 -> 302 | Improved |
| add composition rule to LogicENN_R (ablation) | LogicENN_R · FB15k (raw) | MR: 320 -> 303 | Improved |
| add all rules (LogicENN_R*) instead of no rules (ablation) | LogicENN_R · FB15k (raw) | Hits@10: about 40% -> around 67% | Improved |
| encode rules in ComplEx via RUGE | ComplEx · FB15k (raw) | Hits@10: 48.5 -> 55.5 | Improved |
| use ReLU instead of sigmoid hidden activations (ablation) | LogicENN · FB15k (raw, no rules) | Hits@10: 36.6 -> 42.2 | Improved |
| compare LogicENN_R* against non-rule baselines on rule-light data | LogicENN_R* · FB15k-237 (raw) | Hits@10: 33.3 -> 34.7 | Improved |
| LogicENN_R* versus RotatE on FB15k-237 filtered | LogicENN_R* · FB15k-237 (filtered) | Hits@10: 47.1 -> 47.3 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- RotatE — RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space (ICLR 2019) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Knowledge Graph Embedding With Iterative Guidance From Soft Rules (AAAI 2018 · similarity 0.90)
- Neural Methods for Logical Reasoning over Knowledge Graphs (ICLR 2022 · similarity 0.89)
- Probabilistic Logic Neural Networks for Reasoning (NeurIPS 2019 · similarity 0.89)
- Improving Soft Unification with Knowledge Graph Embedding Methods (ICML 2025 · similarity 0.88)
- Rule-Guided Graph Neural Networks for Explainable Knowledge Graph Reasoning (AAAI 2025 · similarity 0.88)
- Systematic Relational Reasoning With Epistemic Graph Neural Networks (ICLR 2025 · similarity 0.87)
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.
