Keep It Unbiased: A Comparison Between Estimation of Distribution Algorithms and Deep Learning for Human Interaction-Free Side-Channel Analysis
This work addresses the problem of reducing human dependency in SCA security evaluation for cryptographic systems, presenting an incremental comparison between two emerging methods.
The paper compares Estimation of Distribution Algorithms (EDA) and Deep Learning (DL) for human interaction-free side-channel analysis (SCA), finding that EDA offers a simpler alternative without the complexity of neural networks, based on experiments across multiple datasets.
Evaluating side-channel analysis (SCA) security is a complex process, involving applying several techniques whose success depends on human engineering. Therefore, it is crucial to avoid a false sense of confidence provided by non-optimal (failing) attacks. Different alternatives have emerged lately trying to mitigate human dependency, among which deep learning (DL) attacks are the most studied today. DL promise to simplify the procedure by e.g. evading the need for point of interest selection or the capability of bypassing noise and desynchronization, among other shortcuts. However, including DL in the equation comes at a price, since working with neural networks is not straightforward in this context. Recently, an alternative has appeared with the potential to mitigate this dependence without adding extra complexity: Estimation of Distribution Algorithm-based SCA. In this paper, we compare these two relevant methods, supporting our findings by experiments on various datasets.