RAPID: Retrieval Augmented Training of Differentially Private Diffusion Models
This work addresses practical deployment challenges for privacy-preserving generative models, offering a novel approach to enhance efficiency and performance in sensitive data applications.
The paper tackles the problem of utility loss, high memory, and expensive inference in differentially private diffusion models (DPDMs) by introducing RAPID, which integrates retrieval-augmented generation into training, resulting in significant improvements in generative quality, memory footprint, and inference cost over state-of-the-art methods under the same privacy guarantees.
Differentially private diffusion models (DPDMs) harness the remarkable generative capabilities of diffusion models while enforcing differential privacy (DP) for sensitive data. However, existing DPDM training approaches often suffer from significant utility loss, large memory footprint, and expensive inference cost, impeding their practical uses. To overcome such limitations, we present RAPID: Retrieval Augmented PrIvate Diffusion model, a novel approach that integrates retrieval augmented generation (RAG) into DPDM training. Specifically, RAPID leverages available public data to build a knowledge base of sample trajectories; when training the diffusion model on private data, RAPID computes the early sampling steps as queries, retrieves similar trajectories from the knowledge base as surrogates, and focuses on training the later sampling steps in a differentially private manner. Extensive evaluation using benchmark datasets and models demonstrates that, with the same privacy guarantee, RAPID significantly outperforms state-of-the-art approaches by large margins in generative quality, memory footprint, and inference cost, suggesting that retrieval-augmented DP training represents a promising direction for developing future privacy-preserving generative models. The code is available at: https://github.com/TanqiuJiang/RAPID