Explain EEG-based End-to-end Deep Learning Models in the Frequency Domain

TL;DR

This paper introduces a novel explanation method for EEG-based end-to-end deep learning models in the frequency domain, enhancing transparency and understanding of model behavior.

Contribution

We propose a mask perturbation approach with a target alignment loss and a frequency domain perturbation generator to explain EEG models.

Findings

The method effectively explains model behavior in the frequency domain.

Experimental results show the approach's superiority over existing methods.

The approach enhances interpretability of EEG-based deep learning models.

Abstract

The recent rise of EEG-based end-to-end deep learning models presents a significant challenge in elucidating how these models process raw EEG signals and generate predictions in the frequency domain. This challenge limits the transparency and credibility of EEG-based end-to-end models, hindering their application in security-sensitive areas. To address this issue, we propose a mask perturbation method to explain the behavior of end-to-end models in the frequency domain. Considering the characteristics of EEG data, we introduce a target alignment loss to mitigate the out-of-distribution problem associated with perturbation operations. Additionally, we develop a perturbation generator to define perturbation generation in the frequency domain. Our explanation method is validated through experiments on multiple representative end-to-end deep learning models in the EEG decoding field, using an established EEG benchmark dataset. The results demonstrate the effectiveness and superiority of our method, and highlight its potential to advance research in EEG-based end-to-end models.