Research map / Graph learning / Graph self-supervised learning
Evolving Graph Learning for Out-of-Distribution Generalization in Non-Stationary Environments
Qingyun Sun, Jiayi Luo, Haonan Yuan, Xingcheng Fu, Hao Peng, Jianxin Li, Philip S. Yu
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 EvoGOOD instead of EAGLE under node-feature shift (p=0.8) | EvoGOOD · COLLAB node-feature shift (p=0.8) | AUC test: 82.30 -> 93.01 | Improved |
| use EvoGOOD instead of EAGLE under node-feature shift (p=0.4) | EvoGOOD · COLLAB node-feature shift (p=0.4) | AUC test: 88.32 -> 93.12 | Improved |
| use EvoGOOD instead of EAGLE under link-attribute shift | EvoGOOD · Yelp link-attribute shift | AUC w/ OOD: 77.26 -> 78.74 | Improved |
| use EvoGOOD instead of EAGLE under link-attribute shift | EvoGOOD · COLLAB link-attribute shift | AUC w/ OOD: 84.41 -> 84.46 | No clear effect |
| add ESVAE to EAGLE (ablation) | EAGLE · Yelp link-attribute shift | AUC w/ OOD: 77.26 -> 77.78 | Improved |
| add ESVAE to EAGLE (ablation) | EAGLE · COLLAB link-attribute shift | AUC w/ OOD: 84.41 -> 84.45 | No clear effect |
| add ESVAE to EAGLE (ablation) | EAGLE · ACT link-attribute shift | AUC w/ OOD: 82.70 -> 83.20 | Improved |
| add invariant pattern recognition (IPR) to EAGLE (ablation) | EAGLE · COLLAB link-attribute shift | AUC w/ OOD: 84.41 -> 84.09 | No clear effect |
| add invariant pattern recognition (IPR) to EAGLE (ablation) | EAGLE · Yelp link-attribute shift | AUC w/ OOD: 77.26 -> 77.64 | Improved |
| add invariant pattern recognition (IPR) to EAGLE (ablation) | EAGLE · ACT link-attribute shift | AUC w/ OOD: 82.70 -> 83.26 | Improved |
| add ESVAE (sequential environment VAE) instead of a shared VAE (ablation) | EvoGOOD · link prediction (all datasets) | AUC | Improved |
| add causal intervention (L_risk) (ablation) | EvoGOOD · Yelp link prediction | AUC | Improved |
Closest papers by meaning
- Improving Generalization of Dynamic Graph Learning via Environment Prompt (NeurIPS 2024 · similarity 0.91)
- Improving Out-of-Distribution Generalization in Graphs via Hierarchical Semantic Environments (CVPR 2024 · similarity 0.89)
- Graph Invariant Learning with Subgraph Co-mixup for Out-of-Distribution Generalization (AAAI 2024 · similarity 0.89)
- Enhancing Graph Invariant Learning from a Negative Inference Perspective (ICML 2025 · similarity 0.88)
- Temporal Generalization Estimation in Evolving Graphs (ICLR 2024 · similarity 0.88)
- Dissecting the Failure of Invariant Learning on Graphs (NeurIPS 2024 · similarity 0.88)
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.
