AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing

Théo Ryffel; Pierre Tholoniat; David Pointcheval; Francis Bach

AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing

Authors: Théo Ryffel (INRIA, Département d’informatique de l’ENS, ENS, CNRS, PSL University, Paris, France), Pierre Tholoniat (Columbia University, New York, USA), David Pointcheval (Département d’informatique de l’ENS, ENS, CNRS, PSL University, INRIA, Paris, France), Francis Bach (INRIA, Département d’informatique de l’ENS, ENS, CNRS, PSL University, Paris, France)

Volume: 2022
Issue: 1
Pages: 291–316
DOI: https://doi.org/10.2478/popets-2022-0015

Download PDF

Abstract: We propose AriaNN, a low-interaction privacy-preserving framework for private neural network training and inference on sensitive data. Our semi-honest 2-party computation protocol (with a trusted dealer) leverages function secret sharing, a recent lightweight cryptographic protocol that allows us to achieve an efficient online phase. We design optimized primitives for the building blocks of neural networks such as ReLU, MaxPool and BatchNorm. For instance, we perform private comparison for ReLU operations with a single message of the size of the input during the online phase, and with preprocessing keys close to 4× smaller than previous work. Last, we propose an extension to support n-party private federated learning. We implement our framework as an extensible system on top of PyTorch that leverages CPU and GPU hardware acceleration for cryptographic and machine learning operations. We evaluate our end-to-end system for private inference between distant servers on standard neural networks such as AlexNet, VGG16 or ResNet18, and for private training on smaller networks like LeNet. We show that computation rather than communication is the main bottleneck and that using GPUs together with reduced key size is a promising solution to overcome this barrier.

Keywords: Multi Party Computation, Function Secret Sharing, Secure Comparison, Deep Learning

Copyright in PoPETs articles are held by their authors. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 license.