or sheffet

algorithms & theory | machine learning | privacy

Or's work sits at the intersection between machine learning and theory, spanning fields such as algorithms, clustering, algorithmic game theory and recently -- differential privacy, a mathematically rigorous notion of preserving privacy in data analysis. His methods deal with the precise quantification of the inherent trade-offs made between privacy and utility.

Using carefully calibrated random noise, Or’s research can be used to preserve a high-level of individual privacy while providing accurate population-level statistics, estimators and predictions.