FedLeak reconstructs high-fidelity images from federated learning gradients at practical batch sizes, without auxiliary data, by matching only the largest gradient components and regularizing the optimization.
Title resolution pending
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
-
Boosting Gradient Leakage Attacks: Data Reconstruction in Realistic FL Settings
FedLeak reconstructs high-fidelity images from federated learning gradients at practical batch sizes, without auxiliary data, by matching only the largest gradient components and regularizing the optimization.