FedFit paper proposes Fisher Information scoring for federated dynamic sparse training
A new arXiv paper introduces FedFit, a method using Fisher Information scoring to enable dynamic sparse training in cross-device federated learning, addressing memory and communication bottlenecks of training deep neural networks.
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-11-17 05:00 · arXiv cs.LG
FedFit: Federated Dynamic Sparse Training via Fisher Information scoring