Understanding information leakage of distributed inference with deep neural networks: Overview of information theoretic approach and initial results

File Description SizeFormat 
Info-leakage-ML-2018.pdfPublished version666.88 kBAdobe PDFView/Open
Title: Understanding information leakage of distributed inference with deep neural networks: Overview of information theoretic approach and initial results
Authors: Tuor, T
Wang, S
Leung, KK
Ko, BJ
Item 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.
Editors: Kolodny, MA
Wiegmann, DM
Pham, T
Issue Date: 15-Apr-2018
Date of Acceptance: 15-Apr-2018
URI: http://hdl.handle.net/10044/1/69238
DOI: https://doi.org/10.1117/12.2306000
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/Funder: IBM United Kingdom Ltd
Funder's Grant Number: 4603317662
Conference Name: 9th Conference on Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR part of the SPIE Defense + Commercial Sensing Conference
Keywords: 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
Conference Place: Orlando, FL
Online Publication Date: 2018-04-15
Appears in Collections:Computing
Electrical and Electronic Engineering



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx