Abstract:Deep learning models are often trained on datasets that contain sensitive information such as individuals' shopping transactions, personal contacts, and medical records. An increasingly important line of work therefore has sought to train neural networks subject to privacy constraints that are specified by differential privacy or its divergence-based relaxations. These privacy definitions, however, have weaknesses in handling certain important primitives (composition and subsampling), thereby giving loose or complicated privacy analyses of training neural networks. In this paper, we consider a recently proposed privacy definition termed \textit{$f$-differential privacy} [18] for a refined privacy analysis of training neural networks. Leveraging the appealing properties of $f$-differential privacy in handling composition and subsampling, this paper derives analytically tractable expressions for the privacy guarantees of both stochastic gradient descent and Adam used in training deep neural networks, without the need of developing sophisticated techniques as [3] did. Our results demonstrate that the $f$-differential privacy framework allows for a new privacy analysis that improves on the prior analysis~[3], which in turn suggests tuning certain parameters of neural networks for a better prediction accuracy without violating the privacy budget. These theoretically derived improvements are confirmed by our experiments in a range of tasks in image classification, text classification, and recommender systems. Python code to calculate the privacy cost for these experiments is publicly available in the \texttt{TensorFlow Privacy} library.

Differential privacy in deep learning: Privacy and beyond

Differential privacy in deep learning: A literature survey

When Deep Learning Meets Differential Privacy: Privacy,Security, and More

Differential Privacy for Deep and Federated Learning: A Survey

A Survey on Differential Privacy with Machine Learning and Future Outlook

Deep Learning with Differential Privacy

Differential Privacy and Machine Learning: a Survey and Review

More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligence

Differential Privacy in Privacy-Preserving Big Data and Learning: Challenge and Opportunity

Adaptive Differential Privacy Mechanism Based on Entropy Theory for Preserving Deep Neural Networks

Differentially Private Natural Language Models: Recent Advances and Future Directions

Advances in Differential Privacy and Differentially Private Machine Learning

Differential Privacy and Fairness in Decisions and Learning Tasks: A Survey

Privacy at a Price: Exploring its Dual Impact on AI Fairness

An Adaptive and Fast Convergent Approach to Differentially Private Deep Learning

Advancing Differential Privacy: Where We Are Now and Future Directions for Real-World Deployment

Local Differential Privacy and Its Applications: A Comprehensive Survey

Research on the Application of Deep Learning-based Differential Privacy Protection Models in Financial Big Data

Deep Learning with Gaussian Differential Privacy

Privacy Inference Attacks and Defenses in Cloud-based Deep Neural Network: A Survey

A survey on privacy inference attacks and defenses in cloud-based Deep Neural Network