Ph.D. student INSA Lyon, France The CITI Lab at INSA Lyon in France is looking for a PhD student to carry out cutting-edge research in privacy-preserving Federated Learning (FL). FL enables collaborative model training without sharing raw data, preserving privacy by exchanging model updates instead. However, FL remains vulnerable to privacy leakage, poisoning attacks, and challenges from client heterogeneity. Secure Aggregation techniques, such as Homomorphic Encryption, improve privacy, while defenses like anomaly detection and robust aggregation enhance security but often increase computational costs. Asynchronous FL (AsyncFL) improves scalability by processing updates as they arrive, and Buffered AsyncFL helps maintain privacy by aggregating updates in batches. Despite its benefits, FL can be energy-intensive, motivating sustainable approaches such as fog computing and communication-efficient protocols. The proposed SURPRISA-FL framework addresses these challenges by combining priva...