Privacy without Noisy Gradients: Slicing Mechanism for Generative Model Training
This method addresses privacy concerns in generative model training for data scientists by providing a flexible approach that avoids hyper-parameter tuning and convergence issues associated with noisy gradients.
The paper tackles the problem of training generative models with differential privacy by introducing a slicing mechanism that injects noise into random low-dimensional projections of private data, avoiding noisy gradients and enabling higher-quality synthetic data generation compared to baselines.
Training generative models with differential privacy (DP) typically involves injecting noise into gradient updates or adapting the discriminator's training procedure. As a result, such approaches often struggle with hyper-parameter tuning and convergence. We consider the slicing privacy mechanism that injects noise into random low-dimensional projections of the private data, and provide strong privacy guarantees for it. These noisy projections are used for training generative models. To enable optimizing generative models using this DP approach, we introduce the smoothed-sliced $f$-divergence and show it enjoys statistical consistency. Moreover, we present a kernel-based estimator for this divergence, circumventing the need for adversarial training. Extensive numerical experiments demonstrate that our approach can generate synthetic data of higher quality compared with baselines. Beyond performance improvement, our method, by sidestepping the need for noisy gradients, offers data scientists the flexibility to adjust generator architecture and hyper-parameters, run the optimization over any number of epochs, and even restart the optimization process -- all without incurring additional privacy costs.