Sending the locally least-trained model partition to a random neighbor matches or outperforms token-based partitioned gossip learning in complete-graph simulations, with no tokens or metadata.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Stream Learning: Partition-Fair Gossip Learning Without Tokens
Sending the locally least-trained model partition to a random neighbor matches or outperforms token-based partitioned gossip learning in complete-graph simulations, with no tokens or metadata.