Secure Federated Learning against Malicious Attacks

The project focuses on enhancing the security and robustness of federated learning (FL) systems in the presence of malicious actors. Federated learning is a decentralized machine learning approach where multiple participants (e.g., devices, organizations) collaboratively train a shared model without sharing their raw data. While FL preserves data privacy, it is vulnerable to various attacks, such as data poisoning, model poisoning, and inference attacks, which can compromise the integrity and performance of the global model. The project aims to develop techniques and frameworks to detect, mitigate, and prevent such attacks, ensuring the security and reliability of FL systems.

The project has the potential to make FL systems more secure, reliable, and trustworthy. This could enable broader adoption of FL in sensitive applications, such as personalized healthcare, fraud detection, and smart cities, where data privacy and security are paramount. The project contributes to the growing field of secure and privacy-preserving machine learning, offering solutions that protect both the integrity of the global model and the privacy of participants' data.