Fed-REACT first trains a shared encoder, then repeatedly clusters clients by smoothed task-model weights, improving federated learning accuracy on heterogeneous, non-stationary time series.
Adaptive evolutionary clustering
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Fed-REACT: Federated Representation Learning for Heterogeneous and Evolving Data
Fed-REACT first trains a shared encoder, then repeatedly clusters clients by smoothed task-model weights, improving federated learning accuracy on heterogeneous, non-stationary time series.