Research Article
Hyperdimensional Computing for Noninvasive Brain–Computer Interfaces: Blind and One-Shot Classification of EEG Error-Related Potentials
@INPROCEEDINGS{10.4108/eai.22-3-2017.152397, author={Abbas Rahimi and Pentti Kanerva and Jos\^{e} del R. Mill\^{a}n and Jan M. Rabaey}, title={Hyperdimensional Computing for Noninvasive Brain--Computer Interfaces: Blind and One-Shot Classification of EEG Error-Related Potentials}, proceedings={10th EAI International Conference on Bio-inspired Information and Communications Technologies (formerly BIONETICS)}, publisher={EAI}, proceedings_a={BICT}, year={2017}, month={3}, keywords={brain-inspired computing hyperdimensional computing eeg error-related potentials brain-computer interfaces machine learning}, doi={10.4108/eai.22-3-2017.152397} }- Abbas Rahimi
Pentti Kanerva
José del R. Millán
Jan M. Rabaey
Year: 2017
Hyperdimensional Computing for Noninvasive Brain–Computer Interfaces: Blind and One-Shot Classification of EEG Error-Related Potentials
BICT
EAI
DOI: 10.4108/eai.22-3-2017.152397
Abstract
Computing with high-dimensional (HD) vectors, referred to as "hypervectors," is a brain-inspired alternative to computing with numbers. HD computing is characterized by generality, scalability, robustness, and fast learning, making it a prime candidate for utilization in application domains such as brain--computer interfaces. We describe the use of HD computing to classify electroencephalography (EEG) error-related potentials for noninvasive brain--computer interfaces. Our algorithm encodes neural activity recorded from 64 EEG electrodes to a single temporal--spatial hypervector. This hypervector represents the event of interest and is used for recognition of the subject's intentions. Using the full set of training trials, HD computing achieves on average 5% higher accuracy compared to a conventional machine learning method on this task (74.5% vs. 69.5%) and offers further advantages: (1) Our algorithm learns fast by using 34% of training trials while surpassing the conventional method with an average accuracy of 70.5%. (2) Conventional method requires prior domain expert knowledge to carefully select a subset of electrodes for a subsequent preprocessor and classifier, whereas our algorithm blindly uses all 64 electrodes, tolerates noises in data, and the resulting hypervector is intrinsically clustered into HD space; in addition, most preprocessing of the electrode signal can be eliminated while maintaining an average accuracy of 71.7%.
