Saved in:
Bibliographic Details
Main Authors: Jiang, Wuxuan, Song, Xiangjun, Hong, Shenbai, Zhang, Haijun, Liu, Wenxin, Zhao, Bo, Xu, Wei, Li, Yi
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.02320
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914691691839488
author Jiang, Wuxuan
Song, Xiangjun
Hong, Shenbai
Zhang, Haijun
Liu, Wenxin
Zhao, Bo
Xu, Wei
Li, Yi
author_facet Jiang, Wuxuan
Song, Xiangjun
Hong, Shenbai
Zhang, Haijun
Liu, Wenxin
Zhao, Bo
Xu, Wei
Li, Yi
contents Accuracy and efficiency remain challenges for multi-party computation (MPC) frameworks. Spin is a GPU-accelerated MPC framework that supports multiple computation parties and a dishonest majority adversarial setup. We propose optimized protocols for non-linear functions that are critical for machine learning, as well as several novel optimizations specific to attention that is the fundamental unit of Transformer models, allowing Spin to perform non-trivial CNNs training and Transformer inference without sacrificing security. At the backend level, Spin leverages GPU, CPU, and RDMA-enabled smart network cards for acceleration. Comprehensive evaluations demonstrate that Spin can be up to $2\times$ faster than the state-of-the-art for deep neural network training. For inference on a Transformer model with 18.9 million parameters, our attention-specific optimizations enable Spin to achieve better efficiency, less communication, and better accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2402_02320
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Spin: An Efficient Secure Computation Framework with GPU Acceleration
Jiang, Wuxuan
Song, Xiangjun
Hong, Shenbai
Zhang, Haijun
Liu, Wenxin
Zhao, Bo
Xu, Wei
Li, Yi
Cryptography and Security
Machine Learning
Accuracy and efficiency remain challenges for multi-party computation (MPC) frameworks. Spin is a GPU-accelerated MPC framework that supports multiple computation parties and a dishonest majority adversarial setup. We propose optimized protocols for non-linear functions that are critical for machine learning, as well as several novel optimizations specific to attention that is the fundamental unit of Transformer models, allowing Spin to perform non-trivial CNNs training and Transformer inference without sacrificing security. At the backend level, Spin leverages GPU, CPU, and RDMA-enabled smart network cards for acceleration. Comprehensive evaluations demonstrate that Spin can be up to $2\times$ faster than the state-of-the-art for deep neural network training. For inference on a Transformer model with 18.9 million parameters, our attention-specific optimizations enable Spin to achieve better efficiency, less communication, and better accuracy.
title Spin: An Efficient Secure Computation Framework with GPU Acceleration
topic Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2402.02320