Research map / Federated learning and privacy / Personalized and heterogeneous FL
GraphP-FL: Personalized Federated Graph Learning via Dynamic Structure Awareness and Fisher Information Elastic Alignment
Haoyu Chen, Zening Zhao, Jinsong Wang, Kai Shi, Zongpu Wei, Jianhao Li
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 FRPA Fisher elastic alignment (ablation) | GraphP-FL · DAPP | accuracy: 92.22 -> 96.81 | Improved |
| add FRPA Fisher elastic alignment (ablation) | GraphP-FL · NCI1 | accuracy: 89.77 -> 92.55 | Improved |
| add FRPA Fisher elastic alignment (ablation) | GraphP-FL · COLLAB | accuracy: 79.64 -> 87.56 | Improved |
| add contrastive loss GCL (ablation) | GraphP-FL · DAPP | accuracy: 93.69 -> 96.81 | Improved |
| add contrastive loss GCL (ablation) | GraphP-FL · NCI1 | accuracy: 90.56 -> 92.55 | Improved |
| add contrastive loss GCL (ablation) | GraphP-FL · COLLAB | accuracy: 85.45 -> 87.56 | Improved |
| add both FRPA and GCL to FedAvg backbone (ablation) | FedAvg encoder · DAPP | accuracy: 87.77 -> 96.81 | Improved |
| learnable adaptive structural denoising weight instead of fixed (ablation) | GraphP-FL · DD | accuracy: 74.66 -> 82.56 | Improved |
| learnable adaptive structural denoising weight instead of fixed (ablation) | GraphP-FL · MUTAG | accuracy: 80.00 -> 86.25 | Improved |
| learnable adaptive structural denoising weight instead of fixed (ablation) | GraphP-FL · PROTEINS | accuracy: 87.92 -> 90.27 | Improved |
| learnable adaptive structural denoising weight instead of fixed | GraphP-FL · DAPP | accuracy: 96.81 -> 95.79 | Hurt |
| lower Dirichlet alpha from 1.0 to 0.1 | GraphP-FL · DAPP | accuracy: 96.42 -> 95.25 | Hurt |
Closest papers by meaning
- FedSSP: Federated Graph Learning with Spectral Knowledge and Personalized Preference (NeurIPS 2024 · similarity 0.89)
- Federated Graph Learning under Domain Shift with Generalizable Prototypes (AAAI 2024 · similarity 0.89)
- Subgraph Federated Learning via Spectral Methods (NeurIPS 2025 · similarity 0.88)
- Less is More: Federated Graph Learning with Alleviating Topology Heterogeneity from A Causal Perspective (ICML 2025 · similarity 0.88)
- $S^2$FGL: Spatial Spectral Federated Graph Learning (ICML 2025 · similarity 0.87)
- FedSPA: Generalizable Federated Graph Learning under Homophily Heterogeneity (CVPR 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.
