Differentially Private Query Release Through Adaptive Projection
This addresses the challenge of privacy-preserving data analysis for statisticians and data scientists, offering an incremental improvement over prior methods.
The paper tackles the problem of releasing answers to large numbers of statistical queries under differential privacy constraints, proposing an algorithm that adaptively uses a continuous relaxation of the Projection Mechanism with iterative optimization. The method outperforms existing algorithms in many cases, particularly with small privacy budgets or large query classes.
We propose, implement, and evaluate a new algorithm for releasing answers to very large numbers of statistical queries like $k$-way marginals, subject to differential privacy. Our algorithm makes adaptive use of a continuous relaxation of the Projection Mechanism, which answers queries on the private dataset using simple perturbation, and then attempts to find the synthetic dataset that most closely matches the noisy answers. We use a continuous relaxation of the synthetic dataset domain which makes the projection loss differentiable, and allows us to use efficient ML optimization techniques and tooling. Rather than answering all queries up front, we make judicious use of our privacy budget by iteratively and adaptively finding queries for which our (relaxed) synthetic data has high error, and then repeating the projection. We perform extensive experimental evaluations across a range of parameters and datasets, and find that our method outperforms existing algorithms in many cases, especially when the privacy budget is small or the query class is large.