A single malicious client in federated learning can amplify gradient inversion by poisoning its own updates, and common defenses such as robust aggregation or local differential privacy may fail or even increase data leakage.
Trustworthy federated learn- ing: Privacy, security, and beyond,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Byzantine Outside, Curious Inside: Reconstructing Data Through Malicious Updates
A single malicious client in federated learning can amplify gradient inversion by poisoning its own updates, and common defenses such as robust aggregation or local differential privacy may fail or even increase data leakage.