1
How Much Regularization Survives Averaging? Update Masking in Federated Learning
联邦学习中的更新掩码，如何影响正则化在模型平均中的存留？理论剖析与实验验证兼备。
arXiv:2608.23286v1 Announce Type: cross Abstract: Federated learning on non-IID data seeks flat minima to generalize across clients, and existing meth…