Poster: Recovering the Input of Neural Networks via Single Shot Side-channel Attacks

被引：14

作者：

Batina, Lejla ^{[1
]}

Bhasin, Shivam ^{[2
]}

Jap, Dirmanto ^{[2
]}

Picek, Stjepan ^{[3
]}

机构：

[1] Radboud Univ Nijmegen, Nijmegen, Netherlands

[2] Nanyang Technol Univ, Singapore, Singapore

[3] Delft Univ Technol, Delft, Netherlands

来源：

PROCEEDINGS OF THE 2019 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY (CCS'19) | 2019年

基金：

新加坡国家研究基金会;

关键词：

Neural networks; Side-channel analysis; Input recovery;

D O I：

10.1145/3319535.3363280

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The interplay between machine learning and security is becoming more prominent. New applications using machine learning also bring new security risks. Here, we show it is possible to reverse engineer the inputs to a neural network with only a single-shot side-channel measurement assuming the attacker knows the neural network architecture being used.

引用

页码：2657 / 2659

页数：3