When Deep Learning Meets Differential Privacy: Privacy,Security, and More
- 1 November 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 35 (6) , 148-155
Abstract
Over the past decade, we have witnessed unprecedented development in deep learning (DL) and its contributions to modern networking systems. Along with its wide adoption, however, are growing concerns over the broad attack surfaces toward learning systems and the intrinsic vulnerabilities on privacy, security, robustness, and more. As a countermeasure to mitigate the threats or formalize a better defense, a widely adopted approach is to introduce a certain level of random perturbation (a.k.a. calibrated artificial noise) at either the training or prediction phase. Noteworthy examples include effective defenses against model inference attacks and notions of certified robustness. As such, differential privacy (DP), originally established as a privacy-preserving framework for data publishing, has drawn great interest from the learning community. Given a target utility and the acceptable trade-off, DP's formalization on the amount of noise needed has been shown to be widely applicable to a broad range of DL vulnerability mitigations. In this article, we present to our readers the recent representative advancements intersecting DL and DP, ranging from privacy enhancements for DL systems to security and robustness improvements and other novel extensions. Furthermore, we discuss the ongoing challenges and propose a number of future directions where DP has great potential to positively contribute to future DL systems.Keywords
Funding Information
- NSFC (61822309,61773310,U1736205,6213201)
This publication has 8 references indexed in Scilit:
- Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning2021
- Differentially Private and Fair Classification via Calibrated Functional MechanismProceedings of the AAAI Conference on Artificial Intelligence, 2020
- P3SGD: Patient Privacy Preserving SGD for Regularizing Deep CNNs in Pathological Image Classification2019
- Certified Robustness to Adversarial Examples with Differential Privacy2019
- Privacy in Neural Network Learning: Threats and CountermeasuresIEEE Network, 2018
- Privacy Risk in Machine Learning: Analyzing the Connection to Overfitting2018
- Machine Learning Models that Remember Too Much2017
- Deep Learning with Differential Privacy2016