Paper 2023/006

Exploring multi-task learning in the context of two masked AES implementations

Thomas Marquet, Universität Klagenfurt
Elisabeth Oswald, Universität Klagenfurt
Abstract

This paper investigates different ways of applying multi-task learning in the context of two masked AES implementations (via the ASCAD-r and ASCAD-v2 databases). Enabled by multi-task learning, we propose novel architectures that significantly increase the consistency and performance of deep neural networks in a context where the attacker can not access the randomness of the countermeasures during profiling. Our work provides a wide range of experiments to understand the benefits of multi-task strategies against the current single-task state of the art. We show that multi-task learning is significantly more performant than single-task models on all our experiments. Furthermore, such strategies achieve novel milestones against protected implementations as we propose a new best attack on ASCAD-r and ASCAD-v2, along with models that defeat for the first time all masks of the affine masking on ASCAD-v2.

Metadata
Available format(s)
PDF
Category
Attacks and cryptanalysis
Publication info
Preprint.
Keywords
Side Channel AttacksMaskingDeep LearningMulti-task Learning
Contact author(s)
thomas marquet @ aau at
elisabeth oswald @ aau at
History
2023-09-10: last of 3 revisions
2023-01-02: received
See all versions
Short URL
https://ia.cr/2023/006
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2023/006,
      author = {Thomas Marquet and Elisabeth Oswald},
      title = {Exploring multi-task learning in the context of two masked AES implementations},
      howpublished = {Cryptology ePrint Archive, Paper 2023/006},
      year = {2023},
      note = {\url{https://eprint.iacr.org/2023/006}},
      url = {https://eprint.iacr.org/2023/006}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.