Holistic Semi-Supervised Approaches for EEG Representation Learning

TL;DR

This paper adapts and evaluates three state-of-the-art semi-supervised learning methods, originally designed for computer vision, for EEG representation learning, demonstrating their effectiveness with minimal labeled data.

Contribution

It introduces the adaptation of MixMatch, FixMatch, and AdaMatch for EEG data, providing a comprehensive comparison with classical semi-supervised methods on emotion recognition datasets.

Findings

Holistic semi-supervised methods perform well with only 1 labeled sample per class.

AdaMatch generally outperforms other methods in EEG learning.

Semi-supervised approaches significantly improve EEG classification accuracy with limited labels.

Abstract

Recently, supervised methods, which often require substantial amounts of class labels, have achieved promising results for EEG representation learning. However, labeling EEG data is a challenging task. More recently, holistic semi-supervised learning approaches, which only require few output labels, have shown promising results in the field of computer vision. These methods, however, have not yet been adapted for EEG learning. In this paper, we adapt three state-of-the-art holistic semi-supervised approaches, namely MixMatch, FixMatch, and AdaMatch, as well as five classical semi-supervised methods for EEG learning. We perform rigorous experiments with all 8 methods on two public EEG-based emotion recognition datasets, namely SEED and SEED-IV. The experiments with different amounts of limited labeled samples show that the holistic approaches achieve strong results even when only 1 labeled sample is used per class. Further experiments show that in most cases, AdaMatch is the most effective method, followed by MixMatch and FixMatch.