A Differentially Private Framework for Deep Learning with Convexified Loss Functions

Abstract

Differential privacy (DP) has been applied in deep learning for preserving privacy of the underlying training sets. Existing DP practice falls into three categories - objective perturbation, gradient perturbation and output perturbation. They suffer from three main problems. First, conditions on objective functions limit objective perturbation in general deep learning tasks. Second, gradient perturbation does not achieve a satisfactory privacy-utility trade-off due to over-injected noise in each epoch. Third, high utility of the output perturbation method is not guaranteed because of the loose upper bound on the global sensitivity of the trained model parameters as the noise scale parameter. To address these problems, we analyse a tighter upper bound on the global sensitivity of the model parameters. Under a black-box setting, based on this global sensitivity, to control the overall noise injection, we propose a novel output perturbation framework by injecting DP noise into a randomly sampled neuron (via the exponential mechanism) at the output layer of a baseline non-private neural network trained with a convexified loss function. We empirically compare the privacy-utility trade-off, measured by accuracy loss to baseline non-private models and the privacy leakage against black-box membership inference (MI) attacks, between our framework and the open-source differentially private stochastic gradient descent (DP-SGD) approaches on six commonly used real-world datasets. The experimental evaluations show that, when the baseline models have observable privacy leakage under MI attacks, our framework achieves a better privacy-utility trade-off than existing DP-SGD implementations, given an overall privacy budget $\epsilon \le 1$ for a large number of queries.

arXiv commentThis paper has been accepted by the IEEE Transactions on Information Forensics & Security. Early access of IEEE Explore will be available soon

References in corpus (9)

• Deep Learning with Differential Privacy
• Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning
• Renyi Differential Privacy
• Differentially Private Empirical Risk Minimization
• Scalable Private Learning with PATE
• Lipschitz regularity of deep neural networks: analysis and efficient estimation
• Gaussian Differential Privacy
• Functional Mechanism: Regression Analysis under Differential Privacy
• Universal Convexification via Risk-Aversion

Cited by in corpus (1)

• On the Privacy Risks of Deploying Recurrent Neural Networks in Machine Learning Models