DFCA decentralizes IFCA-style clustered federated learning: clients keep one model per cluster, train their assigned model locally, and exchange only that model with neighbors via a running average, matching centralized IFCA accuracy within ~1% (MNIST 92.8, EMNIST 85.3, CIFAR-10 81.5, FEMNIST 87.1 r
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
-
DFCA: Decentralized Federated Clustering Algorithm
DFCA decentralizes IFCA-style clustered federated learning: clients keep one model per cluster, train their assigned model locally, and exchange only that model with neighbors via a running average, matching centralized IFCA accuracy within ~1% (MNIST 92.8, EMNIST 85.3, CIFAR-10 81.5, FEMNIST 87.1 r