Wishart Mechanism for Differentially Private Principal Components Analysis
This work addresses privacy concerns in principal component analysis for data analysts, though it is incremental as it builds on existing input perturbation approaches.
The paper tackles the problem of publishing covariance matrices with differential privacy by proposing a Wishart mechanism that preserves positive semi-definiteness, resulting in better utility guarantees compared to the Laplace mechanism and near-optimal bounds with less computational intractability than previous methods.
We propose a new input perturbation mechanism for publishing a covariance matrix to achieve $(ε,0)$-differential privacy. Our mechanism uses a Wishart distribution to generate matrix noise. In particular, We apply this mechanism to principal component analysis. Our mechanism is able to keep the positive semi-definiteness of the published covariance matrix. Thus, our approach gives rise to a general publishing framework for input perturbation of a symmetric positive semidefinite matrix. Moreover, compared with the classic Laplace mechanism, our method has better utility guarantee. To the best of our knowledge, Wishart mechanism is the best input perturbation approach for $(ε,0)$-differentially private PCA. We also compare our work with previous exponential mechanism algorithms in the literature and provide near optimal bound while having more flexibility and less computational intractability.