Sara Saeidian

I am a postdoctoral researcher at Inria Saclay, in the Comète team, supported by a Swedish Research Council (VR) postdoctoral fellowship. My research explores various aspects of trustworthy machine learning, with a particular focus on privacy and fairness. I am especially interested in developing mathematically rigorous frameworks for analyzing the privacy and fairness guarantees of algorithms, and in understanding their fundamental trade-offs.

I received my PhD in February 2024 from KTH Royal Institute of Technology. During my PhD, I introduced a new privacy measure called pointwise maximal leakage (PML). PML belongs to the family of quantitative information flow definitions and is provably more general than differential privacy, since differential privacy admits an equivalent characterization in terms of PML. My current research builds on my PhD work by developing more theoretical tools in the PML framework.