Research map / Graph learning / Graph self-supervised learning
Exponential Family Graph Embeddings
Abdulkadir Celikkanat, Fragkiskos D. Malliaros
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 |
|---|---|---|---|
| Use EFGE-Norm rather than the strongest baseline | Node classification · CiteSeer, 2% training | Micro-F1: 0.451 -> 0.493 | Improved |
| Use EFGE-Norm rather than the strongest baseline | Node classification · Cora, 2% training | Micro-F1: 0.656 -> 0.682 | Improved |
| Use EFGE-Norm rather than the strongest baseline | Node classification · DBLP, 2% training | Micro-F1: 0.577 -> 0.603 | Improved |
| Use exponential-family embeddings rather than the strongest baseline | Link prediction · CiteSeer | AUC: 0.780 -> 0.834 | Improved |
| Use EFGE-Norm rather than the strongest baseline | Link prediction · Cora | AUC: 0.757 -> 0.807 | Improved |
| Use EFGE-Norm rather than the strongest baseline | Link prediction · AstroPh | AUC: 0.971 -> 0.973 | Improved |
| Use EFGE-Norm rather than the strongest baseline | Link prediction · Facebook | AUC: 0.992 -> 0.992 | No clear effect |
| Increase embedding dimension from 32 to 224 (ablation) | EFGE and comparison embedding models · CiteSeer node classification, 50% training | Micro-F1: around 18% | Improved |
| Increase context window size (ablation) | EFGE-Pois and EFGE-Norm · CiteSeer node classification, 50% training | Micro-F1 | Improved |
| Increase context window size (ablation) | EFGE-Bern · CiteSeer node classification, 50% training | Micro-F1 | Hurt |
| Use a small standard deviation, with best reported results at sigma=1 (ablation) | EFGE-Norm · CiteSeer node classification | Micro-F1 | Improved |
Closest papers by meaning
- Asymptotics of Network Embeddings Learned via Subsampling (JMLR 2023 · similarity 0.87)
- Provable Gaussian Embedding with One Observation (NeurIPS 2018 · similarity 0.87)
- Watch Your Step: Learning Node Embeddings via Graph Attention (NeurIPS 2018 · similarity 0.87)
- Residual2Vec: Debiasing graph embedding with random graphs (NeurIPS 2021 · similarity 0.87)
- Continuous-Time Graph Representation with Sequential Survival Process (AAAI 2024 · similarity 0.86)
- Learning Representations for Hierarchies with Minimal Support (NeurIPS 2024 · similarity 0.85)
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.
