Latent Alignment with Deep Set EEG Decoders

TL;DR

This paper introduces Latent Alignment, a novel deep set-based method for EEG transfer learning that improves classification accuracy across various paradigms by applying statistical distribution alignment at later deep learning stages.

Contribution

The paper presents the Latent Alignment method, which won the BEETL competition, and demonstrates its effectiveness over existing domain adaptation techniques in EEG decoding tasks.

Findings

Latent Alignment outperforms other domain adaptation methods in EEG classification.

Aligning distributions at later deep learning stages enhances accuracy.

Class-imbalance affects the effectiveness of statistical alignment.

Abstract

The variability in EEG signals between different individuals poses a significant challenge when implementing brain-computer interfaces (BCI). Commonly proposed solutions to this problem include deep learning models, due to their increased capacity and generalization, as well as explicit domain adaptation techniques. Here, we introduce the Latent Alignment method that won the Benchmarks for EEG Transfer Learning (BEETL) competition and present its formulation as a deep set applied on the set of trials from a given subject. Its performance is compared to recent statistical domain adaptation techniques under various conditions. The experimental paradigms include motor imagery (MI), oddball event-related potentials (ERP) and sleep stage classification, where different well-established deep learning models are applied on each task. Our experimental results show that performing statistical distribution alignment at later stages in a deep learning model is beneficial to the classification accuracy, yielding the highest performance for our proposed method. We further investigate practical considerations that arise in the context of using deep learning and statistical alignment for EEG decoding. In this regard, we study class-discriminative artifacts that can spuriously improve results for deep learning models, as well as the impact of class-imbalance on alignment. We delineate a trade-off relationship between increased classification accuracy when alignment is performed at later modeling stages, and susceptibility to class-imbalance in the set of trials that the statistics are computed on.