ABY2.0: New Efficient Primitives for STPC with Applications to Privacy in Machine Learning (Extended Abstract)

Arpita Patra, Thomas Schneider, Ajith Suresh, Hossein Yalame

Keywords: Privacy-preserving machine learning, Secure computation

TL;DR: An efficient mixed-protocol framework, outperforming the state-of-the-art STPC frameworks for Privacy in Machine Learning applications

Abstract: In this work, we improve semi-honest secure two-party computation (STPC) over rings, specially for privacy-preserving machine learning, with a focus on the efficiency of the online phase. We construct efficient protocols for several privacy-preserving machine learning (PPML) primitives such as scalar product, matrix multiplication, ReLU, and maxpool. The online communication of our scalar product is two ring elements {\em irrespective} of the vector dimension, which is a feature achieved for the first time in PPML literature. We implement and benchmark training and inference of Logistic Regression and Neural Networks over LAN and WAN networks. For training, we improve online runtime (both for LAN and WAN) over SecureML (Mohassel et al., IEEE S\&P'17) in the range $1.5\times$--$6.1\times$, while for inference, the improvements are in the range of $2.5\times$--$754.3\times$.

Paper Under Submission: The paper is NOT under submission at NeurIPS