Loading [a11y]/accessibility-menu.js
Symmetric matrix perturbation for differentially-private principal component analysis | IEEE Conference Publication | IEEE Xplore

Symmetric matrix perturbation for differentially-private principal component analysis


Abstract:

Differential privacy is a strong, cryptographically-motivated definition of privacy that has recently received a significant amount of research attention for its robustne...Show More
Notes: As originally submitted and published there was an error in this document. The authors subsequently provided the following text: "This paper proposes an algorithm for differentially private PCA that adds a random matrix drawn from the Wishart distribution. The Wishart density is positive only on positive semidefinite matrices can therefore not guarantee (\epsilon,0) differential privacy. Indeed, no additive mechanism with support only on positive semidefinite matrices can guarantee (\epsilon,0) differential privacy. Thus the privacy guarantees in Theorem 1 for Algorithm 1 in this paper are incorrect." The original article PDF remains unchanged.

Abstract:

Differential privacy is a strong, cryptographically-motivated definition of privacy that has recently received a significant amount of research attention for its robustness to known attacks. The principal component analysis (PCA) algorithm is frequently used in signal processing, machine learning and statistics pipelines. In this paper, we propose a new algorithm for differentially-private computation of PCA and compare the performance empirically with some recent state-of-the-art algorithms on different data sets. We intend to investigate the performance of these algorithms with varying privacy parameters and database parameters. We show that our proposed algorithm, despite guaranteeing stricter privacy, provides very good utility for different data sets.
Notes: As originally submitted and published there was an error in this document. The authors subsequently provided the following text: "This paper proposes an algorithm for differentially private PCA that adds a random matrix drawn from the Wishart distribution. The Wishart density is positive only on positive semidefinite matrices can therefore not guarantee (\epsilon,0) differential privacy. Indeed, no additive mechanism with support only on positive semidefinite matrices can guarantee (\epsilon,0) differential privacy. Thus the privacy guarantees in Theorem 1 for Algorithm 1 in this paper are incorrect." The original article PDF remains unchanged.
Date of Conference: 20-25 March 2016
Date Added to IEEE Xplore: 19 May 2016
ISBN Information:
Electronic ISSN: 2379-190X
Conference Location: Shanghai, China

References

References is not available for this document.