Evaluating Fast Adaptability of Neural Networks for Brain-Computer Interface

TL;DR

This paper introduces a straightforward evaluation method for assessing how quickly EEG-based neural network classifiers can adapt to new individuals or tasks, emphasizing the importance of fast calibration in real-world BCI applications.

Contribution

It proposes a simple evaluation strategy and demonstrates that layer-normalization with minimal fine-tuning outperforms MAML in rapid adaptability for EEG classifiers.

Findings

Layer-normalization improves CNN adaptability with few fine-tuning steps.

Simple transfer learning achieves fast adaptation across individuals and tasks.

Empirical comparison favors transfer learning over MAML for quick calibration.

Abstract

Electroencephalography (EEG) classification is a versatile and portable technique for building non-invasive Brain-computer Interfaces (BCI). However, the classifiers that decode cognitive states from EEG brain data perform poorly when tested on newer domains, such as tasks or individuals absent during model training. Researchers have recently used complex strategies like Model-agnostic meta-learning (MAML) for domain adaptation. Nevertheless, there is a need for an evaluation strategy to evaluate the fast adaptability of the models, as this characteristic is essential for real-life BCI applications for quick calibration. We used motor movement and imaginary signals as input to Convolutional Neural Networks (CNN) based classifier for the experiments. Datasets with EEG signals typically have fewer examples and higher time resolution. Even though batch-normalization is preferred for Convolutional Neural Networks (CNN), we empirically show that layer-normalization can improve the adaptability of CNN-based EEG classifiers with not more than ten fine-tuning steps. In summary, the present work (i) proposes a simple strategy to evaluate fast adaptability, and (ii) empirically demonstrate fast adaptability across individuals as well as across tasks with simple transfer learning as compared to MAML approach.