Differentially Private Synthetic Data Using KD-Trees
This work addresses privacy-preserving data sharing for researchers and practitioners, offering a scalable and transparent alternative to deep generative models, though it is incremental in building on existing space partitioning techniques.
The paper tackles the problem of generating differentially private synthetic data that preserves the data distribution, proposing algorithms using KD-trees with noise perturbation to achieve utility improvements over prior work, as shown in empirical evaluations and a downstream classification task.
Creation of a synthetic dataset that faithfully represents the data distribution and simultaneously preserves privacy is a major research challenge. Many space partitioning based approaches have emerged in recent years for answering statistical queries in a differentially private manner. However, for synthetic data generation problem, recent research has been mainly focused on deep generative models. In contrast, we exploit space partitioning techniques together with noise perturbation and thus achieve intuitive and transparent algorithms. We propose both data independent and data dependent algorithms for $ε$-differentially private synthetic data generation whose kernel density resembles that of the real dataset. Additionally, we provide theoretical results on the utility-privacy trade-offs and show how our data dependent approach overcomes the curse of dimensionality and leads to a scalable algorithm. We show empirical utility improvements over the prior work, and discuss performance of our algorithm on a downstream classification task on a real dataset.