Self-Supervised Learning for Electroencephalography

doi:10.1109/TNNLS.2022.3190448

Self-Supervised Learning for Electroencephalography

2024-02-01

146

PRE

AI

M

Mohammad Hossein Rafiei *

DOI：10.1109/TNNLS.2022.3190448

Original

Original request for help

Save

Abstract

En 中文

Decades of research have shown machine learning superiority in discovering highly nonlinear patterns embedded in electroencephalography (EEG) records compared with conventional statistical techniques. However, even the most advanced machine learning techniques require relatively large, labeled EEG repositories. EEG data collection and labeling are costly. Moreover, combining available datasets to achieve a large data volume is usually infeasible due to inconsistent experimental paradigms across trials. Self-supervised learning (SSL) solves these challenges because it enables learning from EEG records across trials with variable experimental paradigms, even when the trials explore different phenomena. It aggregates multiple EEG repositories to increase accuracy, reduce bias, and mitigate overfitting in machine learning training. In addition, SSL could be employed in situations where there is limited labeled training data, and manual labeling is costly. This article: 1) provides a brief introduction to SSL; 2) describes some SSL techniques employed in recent studies, including EEG; 3) proposes current and potential SSL techniques for future investigations in EEG studies; 4) discusses the cons and pros of different SSL techniques; and 5) proposes holistic implementation tips and potential future directions for EEG SSL practices.

Keywords:

Electroencephalography

Brain modeling

Data models

Task analysis

Machine learning

Training

Heuristic algorithms

Electroencephalography (EEG)

machine learning

self-supervised learning (SSL)

Self-Supervised Learning for Electroencephalography

Abstract

Journal

Organization