Abstract: Federated learning (FL) is an innovative privacy-preserving machine learning paradigm that enables clients to train a global model without sharing their local data. However, the coexistence ...