Research map / Federated learning and privacy / Personalized and heterogeneous FL
Medical Federated Model With Mixture of Personalized and Shared Components
Yawei Zhao, Qinghe Liu, Pan Liu, Xinwang Liu, Kunlun He
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 pFedNet instead of FedAvg | federated learning · Luna16 delta=1 | test accuracy: 80.50 -> 86.25 | Improved |
| use pFedNet instead of FedRoD | federated learning · Luna16 delta=2 | test accuracy: 79.25 -> 82.74 | Improved |
| use pFedNet instead of FedAvg | federated learning · BraTS2017 lack #1 | IoU: 70.84 -> 70.40 | Hurt |
| use pFedNet instead of FedRep | federated learning · BraTS2017 lack #0 | label IoU: 68.05 -> 71.25 | Improved |
| use pFedNet instead of FedAvg | federated learning · BraTS2017 lack #2 | IoU: 67.82 -> 71.56 | Improved |
| use pFedNet instead of FedProx | federated learning · BraTS2017 lack #0 | IoU: 50.65 -> 70.73 | Improved |
| use pFedNet instead of Ditto | federated learning · CHD delta=3 | accuracy: 67.37 -> 71.36 | Improved |
| use pFedNet instead of FPFC | federated learning · CHD delta=2 | accuracy: 71.67 -> 70.00 | Hurt |
| add CER on top of STC compression (gamma=50) | STC · BraTS2017 | client compression ratio: 28x -> 52x | Improved |
| add CER with larger gamma on top of STC | STC · BraTS2017 | server compression ratio: 7x -> 17x | Improved |
| use larger CER gamma | pFedNet · BraTS2017 | model performance | Hurt |
| use smaller lambda (more personalized model) | pFedNet · Luna16 | test accuracy | Improved |
Closest papers by meaning
- MH-pFLID: Model Heterogeneous personalized Federated Learning via Injection and Distillation for Medical Data Analysis (ICML 2024 · similarity 0.88)
- FedVCK: Non-IID Robust and Communication-Efficient Federated Learning via Valuable Condensed Knowledge for Medical Image Analysis (AAAI 2025 · similarity 0.87)
- dFLMoE: Decentralized Federated Learning via Mixture of Experts for Medical Data Analysis (CVPR 2025 · similarity 0.86)
- Dual-calibrated Co-training Framework for Personalized Federated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.86)
- FedBiP: Heterogeneous One-Shot Federated Learning with Personalized Latent Diffusion Models (CVPR 2025 · similarity 0.86)
- FedCD: Federated Semi-Supervised Learning with Class Awareness Balance via Dual Teachers (AAAI 2024 · similarity 0.85)
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.
