2019 · 60 citations · 47 references
ComputingComputer ScienceComputer SecurityMachine LearningPrivacy EngineeringData ProtectionData SecurityDifferential PrivacyAdversarial ExamplesAdversarial Privacy AttacksAdversarial RobustnessDifferentially Private Machine LearningPrivacy NotionsPrivacy-preserving LearningPrivacy-preserving Machine LearningRobustness
Adversarial examples that fool machine learning models, particularly deep neural networks, have been a topic of intense research interest, with attacks and defenses being developed in a tight back-and-forth. Most past defenses are best effort and have been shown to be vulnerable to sophisticated attacks. Recently a set of certified defenses have been introduced, which provide guarantees of robustness to norm-bounded attacks. However these defenses either do not scale to large datasets or are limited in the types of models they can support. This paper presents the first certified defense that both scales to large networks and datasets (such as Google's Inception network for ImageNet) and applies broadly to arbitrary model types. Our defense, called PixelDP, is based on a novel connection between robustness against adversarial examples and differential privacy, a cryptographically-inspired privacy formalism, that provides a rigorous, generic, and flexible foundation for defense.
47
ImageNet: A large-scale hierarchical image database
Jia Deng, Wei Dong, Richard Socher et al. · 2009 IEEE Conference on Computer Vision and Pattern Recognition · 2009 · 61.8K citations
Rethinking the Inception Architecture for Computer Vision
Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe et al. · 2016 · 30.9K citations
Deep Learning with Differential Privacy
Martı́n Abadi, Andy Chu, Ian Goodfellow et al. · 2016 · 5.9K citations · Full text