Revisiting Privacy-Utility Trade-off for DP Training with Pre-existing Knowledge

TL;DR

This paper introduces DP-Hero, a novel differential privacy framework that uses heterogeneous noise guided by pre-trained models to improve utility in DP training, including federated learning, while maintaining privacy.

Contribution

It proposes a generic heterogeneous noise mechanism leveraging pre-trained model knowledge to optimize privacy-utility trade-offs in DP training.

Findings

DP-Hero improves training accuracy over state-of-the-art methods.

Heterogeneous noise allocation enhances utility without compromising privacy.

The approach extends effectively to federated learning scenarios.

Abstract

Differential privacy (DP) provides a provable framework for protecting individuals by customizing a random mechanism over a privacy-sensitive dataset. Deep learning models have demonstrated privacy risks in model exposure as an established learning model unintentionally records membership-level privacy leakage. Differentially private stochastic gradient descent (DP-SGD) has been proposed to safeguard training individuals by adding random Gaussian noise to gradient updates in the backpropagation. Researchers identify that DP-SGD causes utility loss since the injected homogeneous noise can alter the gradient updates calculated at each iteration. Namely, all elements in the gradient are contaminated regardless of their importance in updating model parameters. In this work, we argue that the utility can be optimized by involving the heterogeneity of the the injected noise. Consequently, we propose a generic differential privacy framework with heterogeneous noise (DP-Hero) by defining a heterogeneous random mechanism to abstract its property. The insight of DP-Hero is to leverage the knowledge encoded in the previously trained model to guide the subsequent allocation of noise heterogeneity, thereby leveraging the statistical perturbation and achieving enhanced utility. Atop DP-Hero, we instantiate a heterogeneous version of DP-SGD, and further extend it to federated training. We conduct comprehensive experiments to verify and explain the effectiveness of the proposed DP-Hero, showing improved training accuracy compared with state-of-the-art works. Broadly, we shed light on improving the privacy-utility space by learning the noise guidance from the pre-existing leaked knowledge encoded in the previously trained model, showing a different perspective of understanding the utility-improved DP training.