.png){kind=icon}
.png){kind=icon}
Return
A general aggregation federated learning intervention algorithm based on do-calculus
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2025.112210.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Dynamic client weighting employs Monte Carlo sampling for data imbalance. • FedLT-CI enhances tail performance via causal inference, preserving head accuracy. • Causal FL cuts aggregation clients, reducing comms overhead, keeps performance. • Plug-and-play FL boosts tail accuracy while preserving head performance.
Keywords:
Dynamic client weighting
Causal inference
Federated learning
Data imbalance
Communication efficiency