An Algorithmic Framework For Differentially Private Data Analysis on Trusted Processors

Joshua Allen, Bolin Ding, Janardhan Kulkarni, Harsha Nori, Olga Ohrimenko, Sergey Yekhanin

Neural Information Processing Systems 

Differential privacy has emerged as the main definition for private data analysis and machine learning. The global model of differential privacy, which assumes that users trust the data collector, provides strong privacy guarantees and introduces small errors in the output.