Eluding Secure Aggregation in Federated Learning via Model Inconsistency
Primary LanguageJupyter Notebook