HyperFL defends against gradient inversion by sharing only hypernetwork-generated weights, achieving DP-level privacy with higher accuracy.
International Journal of Intelligent Systems , 37(11): 9373– 9389
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A New Federated Learning Framework Against Gradient Inversion Attacks
HyperFL defends against gradient inversion by sharing only hypernetwork-generated weights, achieving DP-level privacy with higher accuracy.