FedSSD applies class-level and sample-level credibility weights to the global model's logits when distilling them into local models, reducing client drift and speeding up convergence in non-IID federated learning.
On the conver- gence of fedavg on non-iid data,
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
-
Learning Critically: Selective Self Distillation in Federated Learning on Non-IID Data
FedSSD applies class-level and sample-level credibility weights to the global model's logits when distilling them into local models, reducing client drift and speeding up convergence in non-IID federated learning.