Being Patient and Persistent: Optimizing An Early Stopping Strategy for Deep Learning in Profiled Attacks

TL;DR

This paper introduces an early stopping algorithm for deep learning in side-channel attacks, improving model training efficiency by reliably identifying optimal states with fewer traces through novel guessing entropy estimation.

Contribution

It presents a new early stopping method based on formalized persistence and patience conditions, enhancing convergence efficiency in deep learning for side-channel analysis.

Findings

Model converges with fewer traces.

Improved reliability in recognizing optimal training state.

Efficient guessing entropy estimation implementation.

Abstract

The absence of an algorithm that effectively monitors deep learning models used in side-channel attacks increases the difficulty of evaluation. If the attack is unsuccessful, the question is if we are dealing with a resistant implementation or a faulty model. We propose an early stopping algorithm that reliably recognizes the model's optimal state during training. The novelty of our solution is an efficient implementation of guessing entropy estimation. Additionally, we formalize two conditions, persistence and patience, for a deep learning model to be optimal. As a result, the model converges with fewer traces.