On the Protection of Private Information in Machine Learning Systems: Two Recent Approaches

Abstract

The recent, remarkable growth of machine learning has led to intense interest in the privacy of the data on which machine learning relies, and to new techniques for preserving privacy. However, older ideas about privacy may well remain valid and useful. This note reviews two recent works on privacy in the light of the wisdom of some of the early literature, in particular the principles distilled by Saltzer and Schroeder in the 1970s.

References in corpus (3)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response
• Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data

Cited by in corpus (7)

• Algorithms that Remember: Model Inversion Attacks and Data Protection Law
• Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inference
• When Machine Learning Meets Privacy: A Survey and Outlook
• GAMIN: An Adversarial Approach to Black-Box Model Inversion
• A Federated Learning Approach for Mobile Packet Classification
• An Overview of Privacy in Machine Learning
• Combining Prediction Intervals on Multi-Source Non-Disclosed Regression Datasets