Research map / Federated learning and privacy / Personalized and heterogeneous FL
Personalized Subgraph Federated Learning
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 functional-similarity weighted aggregation and personalized sparse masks instead of FedAvg | FED-PUB · Overlapping and non-overlapping subgraph node classification | accuracy | Helped |
| Add similarity matching to FedAvg (ablation) | FED-PUB · Cora, overlapping subgraphs | accuracy | Helped |
| Add personalized weight masking to FedAvg (ablation) | FED-PUB · Cora, non-overlapping subgraphs | accuracy | Helped |
| Combine similarity matching and personalized weight masking (ablation) | FED-PUB · Cora, overlapping and non-overlapping subgraphs | accuracy | Helped |
| Remove the proximal term (ablation) | FED-PUB · Subgraph node classification with varying local epochs | accuracy | Hurt |
| Increase local training to five epochs rather than one (ablation) | FED-PUB · Subgraph node classification | accuracy | Hurt |
| Increase the sparsity regularization coefficient λ1 (ablation) | FED-PUB · Cora, overlapping, 10 clients | accuracy | Hurt |
| Use small proximal regularization coefficient λ2 (ablation) | FED-PUB · Cora, overlapping, 10 clients | accuracy | Hurt |
| Use parameter or gradient similarity instead of functional similarity (ablation) | FED-PUB · Subgraph similarity estimation | similarity effectiveness | Hurt |
Closest papers by meaning
- Personalized Subgraph Federated Learning with Differentiable Auxiliary Projections (NeurIPS 2025 · similarity 0.93)
- Subgraph Federated Learning for Local Generalization (ICLR 2025 · similarity 0.92)
- Modeling Inter-Intra Heterogeneity for Graph Federated Learning (AAAI 2025 · similarity 0.91)
- Decoupled Subgraph Federated Learning (ICLR 2025 · similarity 0.90)
- FedSSP: Federated Graph Learning with Spectral Knowledge and Personalized Preference (NeurIPS 2024 · similarity 0.89)
- Virtual Nodes Can Help: Tackling Distribution Shifts in Federated Graph Learning (AAAI 2025 · similarity 0.89)
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.