I am pleased to announce that our paper, “AdaBest: Minimizing Client Drift in Federated Learning via Adaptive Bias Estimation,” has been accepted at ECCV 2022.
