VP-NTK: Exploring the Benefits of Visual Prompting in Differentially Private Data Synthesis

Differentially private (DP) synthetic data has become the de facto standard for releasing sensitive data. However, many DP generative models suffer from the low utility of synthetic data, especially for high-resolution images. On the other hand, one of the emerging techniques in parameter efficient fine- tuning (PEFT) is visual prompting (VP), which allows well- trained existing models to be reused for the purpose of adapt- ing to subsequent downstream tasks. In this work, we ex- plore such a phenomenon in constructing captivating genera- tive models with DP constraints. We show that VP in conjunc- tion with DP-NTK, a DP generator that exploits the power of the neural tangent kernel (NTK) in training DP genera- tive models, achieves a significant performance boost, partic- ularly for high-resolution image datasets, with accuracy im- proving from 0.644±0.044 to 0.769. Lastly, we perform abla- tion studies on the effect of different parameters that influence the overall performance of VP-NTK. Our work demonstrates a promising step forward in improving the utility of DP syn- thetic data, particularly for high-resolution images.