A Joint Exponential Mechanism For Differentially Private Top-$k$

Jennifer Gillenwater · Matthew Joseph · andres munoz · Monica Ribero Diaz

Hall E #912

Keywords: [ SA: Privacy-preserving Statistics and Machine Learning ]

Abstract: We present a differentially private algorithm for releasing the sequence of $k$ elements with the highest counts from a data domain of $d$ elements. The algorithm is a "joint" instance of the exponential mechanism, and its output space consists of all $O(d^k)$ length-$k$ sequences. Our main contribution is a method to sample this exponential mechanism in time $O(dk\log(k) + d\log(d))$ and space $O(dk)$. Experiments show that this approach outperforms existing pure differential privacy methods and improves upon even approximate differential privacy methods for moderate $k$.

Chat is not available.