Paper 2020/165

Efficient Solutions of the CHES 2018 AES Challenge Using Deep Residual Neural Networks and Knowledge Distillation on Adversarial Examples

Aron Gohr and Sven Jacob and Werner Schindler

Abstract

This paper has four main goals. First, we show how we solved the CHES 2018 AES challenge in the contest using essentially a linear classifier combined with a SAT solver and a custom error correction method. This part of the paper has previously appeared in a preprint of our own and later as a contribution to a preprint write-up of the solutions by the three winning teams. This solution serves as a baseline for other solutions explored in the paper. Second, we develop a novel deep neural network architecture for side-channel analysis that completely breaks the AES challenge, allowing for fairly reliable key recovery with just a single trace on the unknown-device part of the CHES challenge. This solution significantly improves upon all previously published solutions of the AES challenge, including our baseline linear solution. Third, we consider the question of leakage attribution for both the classifier we used in the challenge and for our deep neural network. Direct inspection of the weight vector of our machine learning model yields a lot of information on the implementation for our linear classifier. For the deep neural network, we test three other strategies (occlusion of traces; inspection of adversarial changes; knowledge distillation) and find that these can yield information on the leakage essentially equivalent to that gained by inspecting the weights of the simpler model. Fourth, we study the properties of adversarially generated side-channel traces for our model. Partly reproducing recent work on useful features in adversarial examples in our application domain, we find that a linear classifier generalizing to an unseen device much better than our linear baseline can be trained using only adversarial examples (fresh random keys, adversarially perturbed traces) for our deep neural network. This gives a new way of extracting human-usable knowledge from a deep side channel model while also yielding insights on adversarial examples in an application domain where relatively few sources of spurious correlations between data and labels exist.