PASNet: Polynomial Architecture Search Framework for Two-party Computation-based Secure Neural Network Deployment

TL;DR

PASNet is a neural architecture search framework designed for efficient, secure two-party computation-based deep learning, significantly reducing latency and energy consumption while maintaining high accuracy.

Contribution

It introduces a novel framework that incorporates cryptographic hardware latency into neural architecture search, optimizing secure neural network deployment on FPGA.

Findings

Achieves 63 ms and 228 ms latency on ImageNet inference, outperforming state-of-the-art systems.

Attains 70.54% and 78.79% accuracy with over 1000x energy efficiency improvements.

Demonstrates the effectiveness of integrating cryptographic hardware considerations into neural architecture search.

Abstract

Two-party computation (2PC) is promising to enable privacy-preserving deep learning (DL). However, the 2PC-based privacy-preserving DL implementation comes with high comparison protocol overhead from the non-linear operators. This work presents PASNet, a novel systematic framework that enables low latency, high energy efficiency & accuracy, and security-guaranteed 2PC-DL by integrating the hardware latency of the cryptographic building block into the neural architecture search loss function. We develop a cryptographic hardware scheduler and the corresponding performance model for Field Programmable Gate Arrays (FPGA) as a case study. The experimental results demonstrate that our light-weighted model PASNet-A and heavily-weighted model PASNet-B achieve 63 ms and 228 ms latency on private inference on ImageNet, which are 147 and 40 times faster than the SOTA CryptGPU system, and achieve 70.54% & 78.79% accuracy and more than 1000 times higher energy efficiency.