FEDMPR, combining magnitude pruning, dropout, and noise injection in local training, reports accuracy gains over standard federated baselines on several image benchmarks, though not consistently in all settings.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Hybrid-Regularized Magnitude Pruning for Robust Federated Learning under Covariate Shift
FEDMPR, combining magnitude pruning, dropout, and noise injection in local training, reports accuracy gains over standard federated baselines on several image benchmarks, though not consistently in all settings.