Can FedFN Normalize Features in Federated Learning for Data Harmony?

In ML By Bot On

Original title: FedFN: Feature Normalization for Alleviating Data Heterogeneity Problem in Federated Learning

Authors: Seongyoon Kim, Gihun Lee, Jaehoon Oh, Se-Young Yun

In the realm of decentralized learning, Federated Learning (FL) shields data while training models collaboratively. But lurking within this innovation lies a challenge: data diversity hampering performance.

Original article: https://arxiv.org/abs/2311.13267