Understanding information leakage of distributed inference with deep neural networks: Overview of information theoretic approach and initial results
File(s)
Published version
Author(s)
Tuor, Tiffany
Wang, Shiqiang
Leung, Kin K
Ko, Bong Jun
Type
Conference Paper
Abstract
With the emergence of Internet of Things (IoT) and edge computing applications, data is often generated by sensors and end users at the network edge, and decisions are made using these collected data. Edge devices often require cloud services in order to perform intensive inference tasks. Consequently, the inference of deep neural network (DNN) model is often partitioned between the edge and the cloud. In this case, the edge device performs inference up to an intermediate layer of the DNN, and offloads the output features to the cloud for the inference of the remaining of the network. Partitioning a DNN can help to improve energy efficiency but also rises some privacy concerns. The cloud platform can recover part of the raw data using intermediate results of the inference task. Recently, studies have also quantified an information theoretic trade-off between compression and prediction in DNNs. In this paper, we conduct a simple experiment to understand to which extent is it possible to reconstruct the raw data given the output of an intermediate layer, in other words, to which extent do we leak private information when sending the output of an intermediate layer to the cloud. We also present an overview of mutual-information based studies of DNN, to help understand information leakage and some potential ways to make distributed inference more secure.
Editor(s)
Kolodny, MA
Wiegmann, DM
Pham, T
Date Issued
2018-04-15
Date Acceptance
2018-04-15
Citation
Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR IX, 2018, 10635
ISSN
0277-786X
Publisher
Proceedings of SPIE
Journal / Book Title
Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR IX
Volume
10635
Copyright Statement
© 2018 SPIE.
Sponsor
IBM United Kingdom Ltd
Identifier
http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000453766700012&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=1ba7043ffcc86c417c072aa74d649202
Grant Number
4603317662
Source
9th Conference on Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR part of the SPIE Defense + Commercial Sensing Conference
Subjects
Science & Technology
Physical Sciences
Optics
Cloud/edge computing
deep neural networks
distributed inference
Internet of Things
information leakage
mutual information
Publication Status
Published
Start Date
2018-04-15
Finish Date
2018-04-19
Coverage Spatial
Orlando, FL
Date Publish Online
2018-04-15