.png){kind=icon}
Return
FedLoop: Personalized Federated Learning with Closed-Loop Feature Rectification
.png){kind=icon}
PRE
AI
DOI：10.1016/j.inffus.2026.104785.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• FedLoop shifts pFL to dynamic, instance-wise feature rectification.
• Feedback loop decouples feature rectification into direction and magnitude.
• FedLoop achieves SOTA accuracy across modalities with negligible overhead.
Keywords:
Personalized Federated Learning
Closed-Loop
Feature Rectification
Non-IID
Journal
IF:
15.5
Papers:
4.2K
Citations:
2.7W
Organization
View more organizations
Cited Papers
No cited papers available