Efficient and Verifiable Privacy-Preserving Convolutional Computation for CNN Inference with Untrusted Clouds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Jinyu, Sun, Xinrong, Tao, Yunting, Ji, Tong, Kong, Fanyu, Yang, Guoqiang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909742659534848
author Lu, Jinyu
Sun, Xinrong
Tao, Yunting
Ji, Tong
Kong, Fanyu
Yang, Guoqiang
author_facet Lu, Jinyu
Sun, Xinrong
Tao, Yunting
Ji, Tong
Kong, Fanyu
Yang, Guoqiang
contents The widespread adoption of convolutional neural networks (CNNs) in resource-constrained scenarios has driven the development of Machine Learning as a Service (MLaaS) system. However, this approach is susceptible to privacy leakage, as the data sent from the client to the untrusted cloud server often contains sensitive information. Existing CNN privacy-preserving schemes, while effective in ensuring data confidentiality through homomorphic encryption and secret sharing, face efficiency bottlenecks, particularly in convolution operations. In this paper, we propose a novel verifiable privacy-preserving scheme tailored for CNN convolutional layers. Our scheme enables efficient encryption and decryption, allowing resource-constrained clients to securely offload computations to the untrusted cloud server. Additionally, we present a verification mechanism capable of detecting the correctness of the results with a success probability of at least $1-\frac{1}{\left|Z\right|}$. Extensive experiments conducted on 10 datasets and various CNN models demonstrate that our scheme achieves speedups ranging $26 \times$ ~ $\ 87\times$ compared to the original plaintext model while maintaining accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2508_12832
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient and Verifiable Privacy-Preserving Convolutional Computation for CNN Inference with Untrusted Clouds
Lu, Jinyu
Sun, Xinrong
Tao, Yunting
Ji, Tong
Kong, Fanyu
Yang, Guoqiang
Cryptography and Security
Machine Learning
The widespread adoption of convolutional neural networks (CNNs) in resource-constrained scenarios has driven the development of Machine Learning as a Service (MLaaS) system. However, this approach is susceptible to privacy leakage, as the data sent from the client to the untrusted cloud server often contains sensitive information. Existing CNN privacy-preserving schemes, while effective in ensuring data confidentiality through homomorphic encryption and secret sharing, face efficiency bottlenecks, particularly in convolution operations. In this paper, we propose a novel verifiable privacy-preserving scheme tailored for CNN convolutional layers. Our scheme enables efficient encryption and decryption, allowing resource-constrained clients to securely offload computations to the untrusted cloud server. Additionally, we present a verification mechanism capable of detecting the correctness of the results with a success probability of at least $1-\frac{1}{\left|Z\right|}$. Extensive experiments conducted on 10 datasets and various CNN models demonstrate that our scheme achieves speedups ranging $26 \times$ ~ $\ 87\times$ compared to the original plaintext model while maintaining accuracy.
title Efficient and Verifiable Privacy-Preserving Convolutional Computation for CNN Inference with Untrusted Clouds
topic Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2508.12832