Research map / Federated learning and privacy / Communication-efficient FL
SpreadGNN: Decentralized Multi-Task Federated Learning for Graph Neural Networks on Molecular Data
Chaoyang He, Emir Ceyani, Keshav Balasubramanian, Murali Annavaram, Salman Avestimehr
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 decentralized SpreadGNN instead of FedAvg | GraphSAGE · SIDER | ROC-AUC: 0.582 -> 0.5873 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GraphSAGE · Tox21 | ROC-AUC: 0.5548 -> 0.585 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GraphSAGE · MUV | ROC-AUC: 0.6578 -> 0.703 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GraphSAGE · QM8 | MAE: 0.02982 -> 0.02824 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GAT · SIDER | ROC-AUC: 0.5857 -> 0.6034 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GAT · Tox21 | ROC-AUC: 0.6035 -> 0.6056 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GAT · MUV | ROC-AUC: 0.709 -> 0.713 | Improved |
| Use decentralized SpreadGNN instead of FedAvg | GAT · QM8 | MAE: 0.0392 -> 0.0315 | Improved |
| Use decentralized SpreadGNN instead of centralized FedGMTL | GraphSAGE · SIDER | ROC-AUC: 0.629 -> 0.5873 | Hurt |
| Use decentralized SpreadGNN instead of centralized FedGMTL | GraphSAGE · Tox21 | ROC-AUC: 0.6664 -> 0.585 | Hurt |
| Use decentralized SpreadGNN instead of centralized FedGMTL | GraphSAGE · MUV | ROC-AUC: 0.6856 -> 0.703 | Improved |
| Use decentralized SpreadGNN instead of centralized FedGMTL | GraphSAGE · QM8 | MAE: 0.03624 -> 0.02824 | Improved |
Closest papers by meaning
- Virtual Nodes Can Help: Tackling Distribution Shifts in Federated Graph Learning (AAAI 2025 · similarity 0.88)
- Decoupled Subgraph Federated Learning (ICLR 2025 · similarity 0.87)
- Towards Effective Federated Graph Foundation Model via Mitigating Knowledge Entanglement (NeurIPS 2025 · similarity 0.87)
- FedGOG: Federated Graph Out-of-Distribution Generalization with Diffusion Data Exploration and Latent Embedding Decorrelation (AAAI 2025 · similarity 0.87)
- Less is More: Federated Graph Learning with Alleviating Topology Heterogeneity from A Causal Perspective (ICML 2025 · similarity 0.87)
- Federated Graph Learning under Domain Shift with Generalizable Prototypes (AAAI 2024 · 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.
