Sara Saeidianمشاهده پروفایل
پژوهشگر
- Privacy
- Information Theory
- Data Security
- +۳ مورد دیگر
Sara Saeidian is a researcher at KTH Royal Institute of Technology's School of Electrical Engineering and Computer Science (EECS), specifically within the Information Science and Engineering department under the Intelligent systems division. She completed her doctoral dissertation titled "Pointwise Maximal Leakage: Robust, Flexible and Explainable Privacy" in 2024, establishing herself as a promising researcher in information-theoretic privacy. Dr. Saeidian's research program centers on developing a comprehensive framework for privacy-preserving systems with three essential criteria: explainability (operationally meaningful privacy guarantees), robustness (resilience against diverse adversaries), and flexibility (applicability across contexts and data types). Her primary contribution is the development and analysis of pointwise maximal leakage (PML) as a privacy measure that quantifies information leakage about a secret variable to a publicly available related variable. Her publication record from 2021-2025 demonstrates a cohesive research trajectory examining PML's theoretical foundations, composition properties, and practical applications. She has established critical relationships between PML and existing privacy notions like differential privacy, while challenging misconceptions about the impossibility of meaningful inferential privacy guarantees. Her work spans theoretical investigations of optimal privacy mechanisms under leakage constraints to practical applications in privacy-preserving machine learning frameworks like PATE. Dr. Saeidian's research has been published in premier venues including IEEE Transactions on Information Theory, IEEE Transactions on Information Forensics and Security, and proceedings of the IEEE International Symposium on Information Theory, reflecting the significance and quality of her contributions to the field of data privacy.








