Rethinking Privacy with Pointwise Maximal Leakage
Abstract
In this talk, I will discuss the problem of privacy-preserving statistical analysis. I will start with an introduction to differential privacy, a key framework in this area. Then, I will present pointwise maximal leakage (PML), a privacy measure that I developed during my PhD studies. PML quantifies the amount of information leaking about a secret when releasing the outcome of a randomized function calculated on the secret. I will draw connections between PML and differential privacy while also highlighting their differences. Additionally, I will discuss an application where private information is sanitized while guaranteeing privacy in the sense of PML. Finally, I will explore open questions, current, and future research directions.
Sara Saeidian is a postdoctoral researcher at Inria Saclay and KTH Royal Institute of Technology. Her research focuses on privacy-preserving frameworks in statistics and machine learning, with a foundation in information theory. She earned her PhD from KTH in February 2024. She has received an international postdoctoral grant from the Swedish Research Council (Vetenskapsrådet) and the Marie Skłodowska-Curie Actions Postdoctoral Fellowship Seal of Excellence.