A loss-trajectory-based clustering method for federated learning, plus a Wasserstein-adjusted cluster metric, reports improved personalization on heterogeneous data.
(21) If we iterate backward until P 0 kj, we obtain the following update P t+1 kj = (1 − α)t+1P 0 kj + tX τ =0 α(1 − α)τ ωt−τ k
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
-
Interaction-Aware Gaussian Weighting for Clustered Federated Learning
A loss-trajectory-based clustering method for federated learning, plus a Wasserstein-adjusted cluster metric, reports improved personalization on heterogeneous data.