Using an MST plus two-color scheduling to gossip models, the authors report up to 8x bandwidth and 4.4x latency improvements over flooding in a 10-node decentralized federated learning testbed.
Communication-efficient learning of deep networks from decentralized data,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Graph-based Gossiping for Communication Efficiency in Decentralized Federated Learning
Using an MST plus two-color scheduling to gossip models, the authors report up to 8x bandwidth and 4.4x latency improvements over flooding in a 10-node decentralized federated learning testbed.