A Many Objective Optimization Approach for Transfer Learning in EEG Classification

TL;DR

This paper introduces a multi-objective optimization method for transfer learning in EEG classification, enabling effective use of single-source data to improve classification accuracy in brain-computer interfaces, especially for individuals with disabilities.

Contribution

It proposes a novel many-objective optimization framework for transfer learning in EEG classification that does not require multiple source subjects, enhancing applicability and independence.

Findings

Improved classification accuracy on BCI datasets.

Effective feature projection enhances transfer learning.

Method is compatible with existing features and classifiers.

Abstract

In Brain-Computer Interfacing (BCI), due to inter-subject non-stationarities of electroencephalogram (EEG), classifiers are trained and tested using EEG from the same subject. When physical disabilities bottleneck the natural modality of performing a task, acquisition of ample training data is difficult which practically obstructs classifier training. Previous works have tackled this problem by generalizing the feature space amongst multiple subjects including the test subject. This work aims at knowledge transfer to classify EEG of the target subject using a classifier trained with the EEG of another unit source subject. A many-objective optimization framework is proposed where optimal weights are obtained for projecting features in another dimension such that single source-trained target EEG classification performance is maximized with the modified features. To validate the approach, motor imagery tasks from the BCI Competition III Dataset IVa are classified using power spectral density based features and linear support vector machine. Several performance metrics, improvement in accuracy, sensitivity to the dimension of the projected space, assess the efficacy of the proposed approach. Addressing single-source training promotes independent living of differently-abled individuals by reducing assistance from others. The proposed approach eliminates the requirement of EEG from multiple source subjects and is applicable to any existing feature extractors and classifiers. Source code is available at http://worksupplements.droppages.com/tlbci.html.