Interpreting Emergent Features in Deep Learning-based Side-channel Analysis

TL;DR

This paper applies mechanistic interpretability to deep learning models in side-channel analysis, revealing how they exploit leakage and enabling a transition from black-box to white-box evaluation of security vulnerabilities.

Contribution

It introduces a method to interpret neural networks in SCA, uncovering learned representations and secret leaks, which enhances understanding and security assessment.

Findings

Mechanistic interpretability scales to realistic SCA settings.

Models can reveal secret masks despite low accuracy.

Interpretability helps identify leakage sources in side-channel traces.

Abstract

Side-channel analysis (SCA) poses a real-world threat by exploiting unintentional physical signals to extract secret information from secure devices. Evaluation labs also use the same techniques to certify device security. In recent years, deep learning has emerged as a prominent method for SCA, achieving state-of-the-art attack performance at the cost of interpretability. Understanding how neural networks extract secrets is crucial for security evaluators aiming to defend against such attacks, as only by understanding the attack can one propose better countermeasures. In this work, we apply mechanistic interpretability to neural networks trained for SCA, revealing \textit{how} models exploit \textit{what} leakage in side-channel traces. We focus on sudden jumps in performance to reverse engineer learned representations, ultimately recovering secret masks and moving the evaluation process from black-box to white-box. Our results show that mechanistic interpretability can scale to realistic SCA settings, even when relevant inputs are sparse, model accuracies are low, and side-channel protections prevent standard input interventions.