Benutzer: Gast  Login
Titel:

Hide and Seek: Using Occlusion Techniques for Side-Channel Leakage Attribution in CNNs

Dokumenttyp:
Konferenzbeitrag
Autor(en):
Schamberger, Thomas and Egger, Maximilian and Tebelmann, Lars
Abstract:
Deep learning-based side-channel analysis has gained popularity due to its relaxed feature engineering effort in contrast to classical profiled side-channel analysis approaches. This however comes at the cost of a reduced explainability of attack results. In this work we propose occlusion techniques for neural network attribution that allow the identification of points-of-interest related to side-channel leakage used by the networks to defeat masking countermeasures. We evaluate results for both...     »
Herausgeber:
Zhou, Jianying and Batina, Lejla and Li, Zengpeng and Lin, Jingqiang and Losiouk, Eleonora and Majumdar, Suryadipta and Mashima, Daisuke and Meng, Weizhi and Picek, Stjepan and Rahman, Mohammad Ashiqur and Shao, Jun and Shimaoka, Masaki and Soremekun, Ezekiel and Su, Chunhua and Teh, Je Sen and Udovenko, Aleksei and Wang, Cong and Zhang, Leo and Zhauniarovich, Yury
Kongress- / Buchtitel:
Applied Cryptography and Network Security Workshops ACNS
Kongress / Zusatzinformationen:
Kyoto, Japan
Verlag / Institution:
Springer Nature Switzerland
Verlagsort:
Cham
Jahr:
2023
Quartal:
2. Quartal
Jahr / Monat:
2023-06
Seiten:
139--158
Print-ISBN:
978-3-031-41181-6
 BibTeX