Active Semi-supervised Transfer Learning (ASTL) for Offline BCI Calibration

TL;DR

This paper introduces ASTL, a novel approach combining active learning, semi-supervised learning, and transfer learning to improve offline BCI calibration with partially unlabeled auxiliary EEG data, outperforming existing methods.

Contribution

It presents the first active semi-supervised transfer learning method for offline BCI calibration that handles unlabeled auxiliary data, enhancing classification performance across subjects and devices.

Findings

ASTL achieves consistently good performance across subjects.

ASTL outperforms state-of-the-art approaches.

Effective with different EEG headsets.

Abstract

Single-trial classification of event-related potentials in electroencephalogram (EEG) signals is a very important paradigm of brain-computer interface (BCI). Because of individual differences, usually some subject-specific calibration data are required to tailor the classifier for each subject. Transfer learning has been extensively used to reduce such calibration data requirement, by making use of auxiliary data from similar/relevant subjects/tasks. However, all previous research assumes that all auxiliary data have been labeled. This paper considers a more general scenario, in which part of the auxiliary data could be unlabeled. We propose active semi-supervised transfer learning (ASTL) for offline BCI calibration, which integrates active learning, semi-supervised learning, and transfer learning. Using a visual evoked potential oddball task and three different EEG headsets, we demonstrate that ASTL can achieve consistently good performance across subjects and headsets, and it outperforms some state-of-the-art approaches in the literature.