Virtual Homogeneity Learning: Defending against Data Heterogeneity in Federated Learning. (arXiv:2206.02465v2 [cs.LG] UPDATED)

In federated learning (FL), model performance typically suffers from client
drift induced by data heterogeneity, and mainstream works focus on correcting
client drift. We propose a different approach named virtual homogeneity
learning (VHL) to directly "rectify" the data heterogeneity. In particular, VHL
conducts FL with a virtual homogeneous dataset crafted to satisfy two
conditions: containing no private information and being separable. The virtual
dataset can be generated from pure noise shared across clients, aiming to
calibrate the features from the …

arxiv data federated learning learning lg virtual