Two-Stage Data Synthesization: A Statistics-Driven Restricted Trade-off between Privacy and Prediction

TL;DR

This paper introduces a novel two-stage data synthesis method that balances privacy and prediction accuracy by combining statistical techniques with a hybrid approach, validated through theoretical analysis and real-world applications.

Contribution

The paper proposes a two-stage synthesis strategy that effectively manages the privacy-prediction trade-off using a hybrid and kernel ridge regression approach, which is a novel combination in data synthesis.

Findings

Achieves a restricted privacy-prediction trade-off.

Guarantees optimal prediction performance.

Demonstrates effectiveness on real-world datasets.

Abstract

Synthetic data have gained increasing attention across various domains, with a growing emphasis on their performance in downstream prediction tasks. However, most existing synthesis strategies focus on maintaining statistical information. Although some studies address prediction performance guarantees, their single-stage synthesis designs make it challenging to balance the privacy requirements that necessitate significant perturbations and the prediction performance that is sensitive to such perturbations. We propose a two-stage synthesis strategy. In the first stage, we introduce a synthesis-then-hybrid strategy, which involves a synthesis operation to generate pure synthetic data, followed by a hybrid operation that fuses the synthetic data with the original data. In the second stage, we present a kernel ridge regression (KRR)-based synthesis strategy, where a KRR model is first trained on the original data and then used to generate synthetic outputs based on the synthetic inputs produced in the first stage. By leveraging the theoretical strengths of KRR and the covariant distribution retention achieved in the first stage, our proposed two-stage synthesis strategy enables a statistics-driven restricted privacy--prediction trade-off and guarantee optimal prediction performance. We validate our approach and demonstrate its characteristics of being statistics-driven and restricted in achieving the privacy--prediction trade-off both theoretically and numerically. Additionally, we showcase its generalizability through applications to a marketing problem and five real-world datasets.