Scalable Differentially Private Data Compression via Diffusion and Stochastic Codes
For practitioners needing to release privatized high-dimensional data (e.g., images), this provides a practical compression method with controllable tradeoffs, addressing a key bottleneck in DP data release.
The paper tackles the problem of compressing high-dimensional data under differential privacy constraints. DP-DiPP, a pipeline combining stochastic codes with diffusion models, achieves 10-30 times better compression than baselines on CIFAR-10 while maintaining comparable privacy and utility.
The ever-increasing collection of personal data has created mounting pressure to develop technologies that protect sensitive aspects of individual identity. Differential privacy (DP) provides a principled framework with strong formal guarantees and has already achieved practical success. However, releasing high-dimensional data, such as images, has remained elusive: releasing uncompressed privatized data requires significant storage. At the same time, no effective data compression scheme exists that can compress high-resolution data with privacy guarantees. We address this challenge with DP-DiPP, a compression pipeline that combines stochastic codes with diffusion models. DP-DiPP is highly flexible: the practitioner has direct control over the compression rate-privacy-utility tradeoff. As the theoretical backbone, we extend the Poisson private representation (PPR) to encode the outputs of privacy mechanisms. We then combine it with DiffC, a diffusion-based lossy data compression method, to obtain a differentially private image compressor. Our experiments on privatized image classification on CIFAR-10 demonstrate that DP-DiPP significantly outperforms the baseline, achieving a 10-30 times better compression while retaining comparable privacy guarantees and utility.