Differential privacy and applications
Differential privacy offers ways to answer statistical queries about sensitive data while providing strong provable privacy guarantees ensuring that the presence or absence of a single individual in the data has a negligible statistical effect on the query's result.
In this class, I will first introduce the basic concepts of differential privacy and some of the fundamental mechanisms for building differentially private programs. Then, I will discuss several recently developed models for differential privacy in a multi-party setting. To conclude, I will discuss the relations of differential privacy with statistics, machine learning, and adaptive data analysis.
- M. Gaboardi. Differential privacy and applications. Slides from EWSCS 2020.
- Videos from the lectures (large, unedited files) [mp4, password-protected]
April 10, 2020 22:30 Europe/Helsinki (GMT +03:00)
local organizers, ewscs20(at)cs.ioc.ee
EWSCS'20 page: //cs.ioc.ee/ewscs/2020/