Research map / Graph learning / Graph self-supervised learning
Learning Invariant Representations of Graph Neural Networks via Cluster Generalization
Donglin Xia, Xiao Wang, Nian Liu, Chuan Shi
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 cluster information transfer to the GNN | APPNP · Cora, Citeseer, and Pubmed, ADD-0.5 | accuracy: 79.54±0.50 -> 80.50±0.39; 66.96±0.76 -> 68.54±0.71; 75.88±0.81 -> 76.64±0.40 | Improved |
| Add cluster information transfer to GNN backbones | GNNs · ACM PLP | accuracy: 82.48±1.73 -> 85.75±1.76 | Improved |
| Add cluster information transfer to GCN | GCN · IMDB MAM | accuracy: 58.98±1.11 -> 56.68±1.45 | Hurt |
| Add cluster information transfer to GNN backbones | GNNs · Multiplex networks, across ACM and IMDB relations | accuracy | Mixed |
| Add Gaussian perturbations to cluster transfer (ablation) | CIT-GNN · IMDB MDM | accuracy: 56.37±1.41 -> 55.86±1.54 | Hurt |
| Add Gaussian perturbations to cluster transfer (ablation) | CIT-GNN · Cora, Citeseer, and Pubmed, ADD-0.5 | accuracy: 76.88±0.34 -> 76.98±0.49; 66.67±0.55 -> 67.65±0.44; 72.83±0.32 -> 73.76±0.40 | Improved |
| Vary transfer probability p (ablation) | CIT-GNN · Hyperparameter analysis | accuracy | Mixed |
| Vary transfer interval k (ablation) | CIT-GNN · Hyperparameter analysis | accuracy | No clear effect |
| Vary the number of clusters m (ablation) | CIT-GNN · Cora and Citeseer | accuracy | Mixed |
Closest papers by meaning
- Learning Invariant Graph Representations for Out-of-Distribution Generalization (NeurIPS 2022 · similarity 0.89)
- Graph Neural Networks are Inherently Good Generalizers: Insights by Bridging GNNs and MLPs (ICLR 2023 · similarity 0.89)
- InterpGNN: Understand and Improve Generalization Ability of Transdutive GNNs through the Lens of Interplay between Train and Test Nodes (ICLR 2024 · similarity 0.89)
- Contextual Structure Knowledge Transfer for Graph Neural Networks (AAAI 2025 · similarity 0.89)
- On the Interplay between Graph Structure and Learning Algorithms in Graph Neural Networks (ICML 2025 · similarity 0.88)
- Learning to Reweight for Generalizable Graph Neural Network (AAAI 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.
