Aegis: A Lightning Fast Privacy-preserving Machine Learning Platform against Malicious Adversaries

ePrint Report: Aegis: A Lightning Fast Privacy-preserving Machine Learning Platform against Malicious Adversaries

Taipei Lu, Bingsheng Zhang, Lichun Li, Kui Ren

Privacy-preserving machine learning (PPML) techniques have gained significant popularity in the past years. Those protocols have been widely adopted in many real-world security-sensitive machine learning scenarios, e.g., medical care and finance. In this work, we introduce $\mathsf{Aegis}$~-- a high-performance PPML platform built on top of a maliciously secure 3-PC framework over ring $\mathbb{Z}_{2^\ell}$. In particular, we propose a novel …

adversaries aegis eprint report fast lightning machine machine learning malicious medical platform privacy protocols real report security sensitive taipei techniques world