TONIC: Towards oblivious neural inference compiler

Po Hsuan Huang, Chia Heng Tu, Shen Ming Chung

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Privacy-preserving deep learning computing becomes popular these days as it helps protect, for example, both user data and deep neural network (DNN) model parameters at the same time with cryptographic techniques. In particular, significant efforts have been made to leverage secure two-party computation schemes for preventing user/model data from disclosure during DNN inference. Nevertheless, the existing works require manual intervention while converting trained models into secure computation programs, which is not scalable to modern deep networks efficiently. In this work, we propose a compiler framework, TONIC, to do the conversion automatically with scalability. Given a pre-trained DNN model, TONIC converts it into one of two secure two-party computation languages, i.e., ObliVM and ABY. Based on tailored backends built on top of a DNN compiler, TVM, our case studies show that TONIC is able to automatically convert popular DNN models, such as CryptoNets and MobileNetV2, into the corresponding programs for secure computations.

Original languageEnglish
Title of host publicationProceedings of the 36th Annual ACM Symposium on Applied Computing, SAC 2021
PublisherAssociation for Computing Machinery
Pages491-500
Number of pages10
ISBN (Electronic)9781450381048
DOIs
Publication statusPublished - 2021 Mar 22
Event36th Annual ACM Symposium on Applied Computing, SAC 2021 - Virtual, Online, Korea, Republic of
Duration: 2021 Mar 222021 Mar 26

Publication series

NameProceedings of the ACM Symposium on Applied Computing

Conference

Conference36th Annual ACM Symposium on Applied Computing, SAC 2021
CountryKorea, Republic of
CityVirtual, Online
Period21-03-2221-03-26

All Science Journal Classification (ASJC) codes

  • Software

Fingerprint Dive into the research topics of 'TONIC: Towards oblivious neural inference compiler'. Together they form a unique fingerprint.

Cite this