Practical Membership Inference Attack Against Collaborative Inference in Industrial IoT

The effectiveness of state-of-the-art deep learning (DL) models has empowered the development of industrial Internet of things (IIoT). Recently, considering resource-constrained and privacy-required IIoT devices, collaborative inference has been proposed, which splits DL models and deploys them in I...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on industrial informatics 2022-01, Vol.18 (1), p.477-487
Main Authors: Chen, Hanxiao, Li, Hongwei, Dong, Guishan, Hao, Meng, Xu, Guowen, Huang, Xiaoming, Liu, Zhe
Format: Article
Language:English
Subjects:
Collaboration
Collaborative inference
Computational modeling
Data models
Data privacy
Deep learning
deep learning (DL)
Industrial applications
industrial Internet of things (IIoT)
Inference
Internet of Things
membership inference attack
Privacy
Servers
Training
Training data
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The effectiveness of state-of-the-art deep learning (DL) models has empowered the development of industrial Internet of things (IIoT). Recently, considering resource-constrained and privacy-required IIoT devices, collaborative inference has been proposed, which splits DL models and deploys them in IIoT devices and an edge server separately. However, in this article, we argue that there are still severe privacy vulnerabilities in collaborative inference systems. And we devise the first membership inference attack (MIA) against collaborative inference, to infer whether a particular data sample is used for training the model of IIoT systems. Existing MIAs either assume full access to the systems' APIs or availability of the target model's parameters, which is not applicable in realistic IIoT environments. In contrast to prior works, we propose transfer-inherit shadow learning and thus relax these key assumptions. We evaluate our attack on different datasets and various settings, and the results show it has high effectiveness.
ISSN:1551-3203
1941-0050
DOI:10.1109/TII.2020.3046648