SecureFed combines PCA-based anomaly detection, validation loss, and gradient magnitude into a trust score that controls zone-weighted aggregation, improving accuracy over vanilla FedAvg in a small MNIST label-flipping experiment.
Towards federated learning at scale: System design,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
SecureFed: A Two-Phase Framework for Detecting Malicious Clients in Federated Learning
SecureFed combines PCA-based anomaly detection, validation loss, and gradient magnitude into a trust score that controls zone-weighted aggregation, improving accuracy over vanilla FedAvg in a small MNIST label-flipping experiment.