PINFER: Privacy-Preserving Inference for Machine Learning

TL;DR

This paper introduces protocols for privacy-preserving machine learning inference that use simple homomorphic encryption, minimize interactions, and are applicable to algorithms like logistic regression, SVM, and neural networks.

Contribution

It presents novel protocols that enable privacy-preserving inference with minimal communication using only additively homomorphic encryption.

Findings

Protocols support logistic regression and SVM classification

Extensions to neural networks demonstrated

Limited interaction protocols reduce communication overhead

Abstract

The foreseen growing role of outsourced machine learning services is raising concerns about the privacy of user data. Several technical solutions are being proposed to address the issue. Hardware security modules in cloud data centres appear limited to enterprise customers due to their complexity, while general multi-party computation techniques require a large number of message exchanges. This paper proposes a variety of protocols for privacy-preserving regression and classification that (i) only require additively homomorphic encryption algorithms, (ii) limit interactions to a mere request and response, and (iii) that can be used directly for important machine-learning algorithms such as logistic regression and SVM classification. The basic protocols are then extended and applied to feed-forward neural networks.