Privacy Analysis

Privacy analysis focuses on quantifying and mitigating the risk of sensitive information leakage in various data-driven systems, particularly those employing machine learning. Current research emphasizes developing tighter privacy guarantees for algorithms like noisy stochastic gradient descent (SGD) used in federated learning and differentially private mechanisms, often exploring techniques like hidden state analysis and advanced composition theorems to improve accuracy while maintaining privacy. These advancements are crucial for building trustworthy AI systems and ensuring compliance with data protection regulations across diverse applications, from healthcare to autonomous vehicles.