Paper 2023/100

Meteor: Improved Secure 3-Party Neural Network Inference with Reducing Online Communication Costs

Ye Dong, Institute of Information Engineering,Chinese Academy of Sciences
Xiaojun Chen, Institute of Information Engineering,Chinese Academy of Sciences
Weizhan Jing, Institute of Information Engineering,Chinese Academy of Sciences
Kaiyun Li, Institute of Information Engineering,Chinese Academy of Sciences
Weiping Wang, Institute of Information Engineering,Chinese Academy of Sciences
Abstract

Secure neural network inference has been a promising solution to private Deep-Learning-as-a-Service, which enables the service provider and user to execute neural network inference without revealing their private inputs. However, the expensive overhead of current schemes is still an obstacle when applied in real applications. In this work, we present \textsc{Meteor}, an online communication-efficient and fast secure 3-party computation neural network inference system aginst semi-honest adversary in honest-majority. The main contributions of \textsc{Meteor} are two-fold: \romannumeral1) We propose a new and improved 3-party secret sharing scheme stemming from the \textit{linearity} of replicated secret sharing, and design efficient protocols for the basic cryptographic primitives, including linear operations, multiplication, most significant bit extraction, and multiplexer. \romannumeral2) Furthermore, we build efficient and secure blocks for the widely used neural network operators such as Matrix Multiplication, ReLU, and Maxpool, along with exploiting several specific optimizations for better efficiency. Our total communication with the setup phase is a little larger than SecureNN (PoPETs'19) and \textsc{Falcon} (PoPETs'21), two state-of-the-art solutions, but the gap is not significant when the online phase must be optimized as a priority. Using \textsc{Meteor}, we perform extensive evaluations on various neural networks. Compared to SecureNN and \textsc{Falcon}, we reduce the online communication costs by up to $25.6\times$ and $1.5\times$, and improve the running-time by at most $9.8\times$ (resp. $8.1\times$) and $1.5\times$ (resp. $2.1\times$) in LAN (resp. WAN) for the online inference.

Metadata
Available format(s)
PDF
Category
Cryptographic protocols
Publication info
Published elsewhere. ACM WWW2023
Keywords
PrivacySecuritySecret SharingNeural Network
Contact author(s)
dongye @ iie ac cn
chenxiaojun @ iie ac cn
History
2023-01-27: approved
2023-01-27: received
See all versions
Short URL
https://ia.cr/2023/100
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2023/100,
      author = {Ye Dong and Xiaojun Chen and Weizhan Jing and Kaiyun Li and Weiping Wang},
      title = {Meteor: Improved Secure 3-Party Neural Network Inference with Reducing Online Communication Costs},
      howpublished = {Cryptology ePrint Archive, Paper 2023/100},
      year = {2023},
      note = {\url{https://eprint.iacr.org/2023/100}},
      url = {https://eprint.iacr.org/2023/100}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.