Poster : recovering the input of neural networks via single shot side-channel attacks

• Main Authors: Batina, Lejla, Jap, Dirmanto, Bhasin, Shivam, Picek, Stjepan
• Other Authors: Conference on Computer and Communications Security (CCS 2019)
• Format: Conference or Workshop Item
• Language: English
• Published: 2021
• Subjects: Science::Mathematics::Discrete mathematics::Cryptography; Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence; Neural Networks; Side-channel Analysis
• Online Access: https://hdl.handle.net/10356/148356
• Institution: Nanyang Technological University

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 attacke...