A malicious federated learning server can sweep bias hyperplanes across training rounds and perfectly reconstruct every sample in a client's full batch, for images and tabular data.
Anguita, Alessandro Ghio, L
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Cutting Through Privacy: A Hyperplane-Based Data Reconstruction Attack in Federated Learning
A malicious federated learning server can sweep bias hyperplanes across training rounds and perfectly reconstruct every sample in a client's full batch, for images and tabular data.