Dr. Or Seffet

Differential privacy


My interests lie in the rapidly-growing field of differential privacy: the ability to conduct data-analysis over sensitive personal data (location, medical history, financial records etc.) while adhering to a rigorous and mathematical notion of privacy that is guaranteed to bound any user’s privacy-loss. The goal of my research is to provide data-analysis techniques and algorithms that work by injecting carefully designed and carefully calibrated noise into the computation, so that on the one hand we obfuscate the presence or absence of any single individual while on the other hand we introduce only limited interference with the result of analysis. I’m particularly interested in differentially private machine learning and statistical inference; yet I also work on more general machine learning theory and approximation algorithms.


I am a Senior Lecturer (Assistant Prof.) in the Faculty of Engineering in Bar-Ilan University since 2019. I received my PhD in computer science from Carnegie Mellon University under the supervision of Prof. Avrim Blum in 2013. Since then I attended the “Theoretical Foundations of Big Data” program at the Simons Institute for the Theory of Computing, UC Berkeley, and was a postdoctoral fellow at Harvard University’s School of Engineering and Applied Sciences (SEAS), as well as a visiting scholar at the University of Ottawa. Between 2016-9 I was an Assistant Professor at the Department of Computing Science at the University of Alberta, as well as a PI of the Alberta Machine Intelligence Institute (AMII), and I also attended the Simons’ Institute program “Data Privacy: Foundations and Applications” during spring 2019.