1 Appendix 1.1 Preliminaries and Related Works 1.1.1 Federated Learning Suppose there are m clients in a FL system, and each client k has its own dataset D

FL also reduces the risk of being attacked, since the communication happens only once. However, they introduced a public dataset to enhance training, which is not practical. Overall, none of the above methods can be practically applied. In traditional FL frameworks, all users have to agree on the specific architecture of the global model. To support model heterogeneity, Li et al. [ Federated learning: Challenges, methods, and future directions.