Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data

Abstract

Some machine learning applications involve training data that is sensitive, such as the medical histories of patients in a clinical trial. A model may inadvertently and implicitly store some of its training data; careful analysis of the model may therefore reveal sensitive information. To address this problem, we demonstrate a generally applicable approach to providing strong privacy guarantees for training data: Private Aggregation of Teacher Ensembles (PATE). The approach combines, in a black-box fashion, multiple models trained with disjoint datasets, such as records from different subsets of users. Because they rely directly on sensitive data, these models are not published, but instead used as "teachers" for a "student" model. The student learns to predict an output chosen by noisy voting among all of the teachers, and cannot directly access an individual teacher or the underlying data or parameters. The student's privacy properties can be understood both intuitively (since no single teacher and thus no single dataset dictates the student's training) and formally, in terms of differential privacy. These properties hold even if an adversary can not only query the student but also inspect its internal workings. Compared with previous work, the approach imposes only weak assumptions on how teachers are trained: it applies to any model, including non-convex models like DNNs. We achieve state-of-the-art privacy/utility trade-offs on MNIST and SVHN thanks to an improved privacy analysis and semi-supervised learning.

Accepted to ICLR 17 as an oral

Cited by in corpus (12)

• Cronus: Robust and Heterogeneous Collaborative Learning with Black-Box Knowledge Transfer
• Generative Models for Effective ML on Private, Decentralized Datasets
• Model extraction from counterfactual explanations
• An Overview of Privacy in Machine Learning
• A Flexible Selection Scheme for Minimum-Effort Transfer Learning
• Differentially Private Deep Learning with Smooth Sensitivity
• Model Compression with Two-stage Multi-teacher Knowledge Distillation for Web Question Answering System
• Differentially Private Continual Learning
• Towards a General Model of Knowledge for Facial Analysis by Multi-Source Transfer Learning
• P3SGD: Patient Privacy Preserving SGD for Regularizing Deep CNNs in Pathological Image Classification
• An Adaptive and Fast Convergent Approach to Differentially Private Deep Learning
• Efficient Private Machine Learning by Differentiable Random Transformations