Welcome to the resource topic for 2025/1917
Title:
Embedding belief propagation within a multi-task learning model : An example on Kyber’s NTT
Authors: Thomas Marquet, Elisabeth Oswald
Abstract:Deep learning has become a powerful tool for profiled side-channel analysis, especially for its ability to defeat masking countermeasures. However, obtaining a successful deep learning model when the attacker cannot access the internal randomness of the profiling device, remains a challenge. The “plateau effect” hinders the training of the model, as the optimization stalls in flat regions of the loss landscape at initialization, which makes the outcome of the training run uncertain. Previous works showed that multi-task learning allows to overcome this problem by leveraging redundant features across multiple tasks, such as shared randomness or common masking flow. We continue the discussion by using belief propagation on a larger graph to guide the learning. We introduce a multi-task learning model that explicitly integrates a factor graph reflecting the algebraic dependencies among intermediates in the computations of Kyber’s inverse Number Theoretic Transform (iNTT). Such framework allow the model to learn a joint representation of the related tasks that is mutually beneficial, and provides a mechanism to overcome such plateaus. For the first time, we show that one can perform a belief propagation during training even when one does not have access to the internal randomness, on the masked shares, potentially improving greatly the performances of the attack.
ePrint: https://eprint.iacr.org/2025/1917
See all topics related to this paper.
Feel free to post resources that are related to this paper below.
Example resources include: implementations, explanation materials, talks, slides, links to previous discussions on other websites.
For more information, see the rules for Resource Topics .