Can differential privacy practically protect collaborative deep learning inference for IoT?

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

5 Scopus Citations
View graph of relations

Author(s)

  • Jihyeon Ryu
  • Yansong Gao
  • Alsharif Abuadbba
  • Junyaup Kim
  • Dongho Won
  • Surya Nepal
  • Hyoungshick Kim

Related Research Unit(s)

Detail(s)

Original languageEnglish
Journal / PublicationWireless Networks
Online published5 Sept 2022
Publication statusOnline published - 5 Sept 2022

Abstract

Collaborative inference has recently emerged as an attractive framework for applying deep learning to Internet of Things (IoT) applications by splitting a DNN model into several subpart models among resource-constrained IoT devices and the cloud. However, the reconstruction attack was proposed recently to recover the original input image from intermediate outputs that can be collected from local models in collaborative inference. For addressing such privacy issues, a promising technique is to adopt differential privacy so that the intermediate outputs are protected with a small accuracy loss. In this paper, we provide the first systematic study to reveal insights regarding the effectiveness of differential privacy for collaborative inference against the reconstruction attack. We specifically explore the privacy-accuracy trade-offs for three collaborative inference models with four datasets (SVHN, GTSRB, STL-10, and CIFAR-10). Our experimental analysis demonstrates that differential privacy can practically be applied to collaborative inference when a dataset has small intra-class variations in appearance. With the (empirically) optimized privacy budget parameter in our study, the differential privacy technique incurs accuracy loss of 0.476%, 2.066%, 5.021%, and 12.454% on SVHN, GTSRB, STL-10, and CIFAR-10 datasets, respectively, while thwarting the reconstruction attack.

Research Area(s)

  • Cloud computing, Collaborative inference, Data reconstruction attack, Differential privacy

Citation Format(s)

Can differential privacy practically protect collaborative deep learning inference for IoT? / Ryu, Jihyeon; Zheng, Yifeng; Gao, Yansong et al.
In: Wireless Networks, 05.09.2022.

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review