FedHAW dynamically adjusts aggregation weights in federated learning via hypergradient descent, yielding better generalization under data heterogeneity and robustness to communication errors.
Revisiting weighted aggregation in Federated Learning with neural networks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Federated Learning with Hypergradient-based Online Update of Aggregation Weights
FedHAW dynamically adjusts aggregation weights in federated learning via hypergradient descent, yielding better generalization under data heterogeneity and robustness to communication errors.