A spectral-ensemble deep random vector functional link network for passive brain–computer interface

<p dir="ltr">Randomized <u>neural networks</u> (RNNs) have shown outstanding performance in many different fields. The superiority of having fewer training parameters and closed-form solutions makes them popular in small datasets analysis. However, automatically decoding...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Ruilin Li (5627456) (author)
مؤلفون آخرون: Ruobin Gao (16003195) (author), Ponnuthurai N. Suganthan (17347024) (author), Jian Cui (182110) (author), Olga Sourina (17823674) (author), Lipo Wang (17185534) (author)
منشور في: 2023
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513545941549056
author Ruilin Li (5627456)
author2 Ruobin Gao (16003195)
Ponnuthurai N. Suganthan (17347024)
Jian Cui (182110)
Olga Sourina (17823674)
Lipo Wang (17185534)
author2_role author
author
author
author
author
author_facet Ruilin Li (5627456)
Ruobin Gao (16003195)
Ponnuthurai N. Suganthan (17347024)
Jian Cui (182110)
Olga Sourina (17823674)
Lipo Wang (17185534)
author_role author
dc.creator.none.fl_str_mv Ruilin Li (5627456)
Ruobin Gao (16003195)
Ponnuthurai N. Suganthan (17347024)
Jian Cui (182110)
Olga Sourina (17823674)
Lipo Wang (17185534)
dc.date.none.fl_str_mv 2023-05-23T15:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.eswa.2023.120279
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/A_spectral-ensemble_deep_random_vector_functional_link_network_for_passive_brain_computer_interface/29445194
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biomedical and clinical sciences
Neurosciences
Engineering
Biomedical engineering
Information and computing sciences
Artificial intelligence
Ensemble deep random vector functional link (edRVFL)
Spectral-edRVFL (SedRVFL)
Electroencephalogram (EEG)
Feature-refining (FR) block
Dynamic direct link (DDL)
dc.title.none.fl_str_mv A spectral-ensemble deep random vector functional link network for passive brain–computer interface
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Randomized <u>neural networks</u> (RNNs) have shown outstanding performance in many different fields. The superiority of having fewer training parameters and closed-form solutions makes them popular in small datasets analysis. However, automatically decoding raw <u>electroencephalogram</u> (EEG) data using RNNs is still challenging in EEG-based passive brain–computer interface (pBCI) <u>classification tasks</u>. Models with the high-dimension input of EEG may suffer from overfitting and the intrinsic characteristics of non-stationary, high-level noises and subject variability could limit the generation of distinctive features in the hidden layers. To address these problems in EEG-based pBCI tasks, this work proposes a spectral-ensemble deep random vector functional link (SedRVFL) network that focuses on feature learning in the frequency domain. Specifically, an unsupervised feature-refining (FR) block is proposed to improve the low <u>feature learning</u> capability in RNNs. Moreover, a dynamic direct link (DDL) is performed to further complement the <u>frequency information</u>. The proposed model has been evaluated on a self-collected dataset as well as a public driving dataset. The cross-subject <u>classification results</u> obtained demonstrated its effectiveness. This work offers a new solution for EEG decoding, i.e., using optimized RNNs for decoding complex raw EEG data and boosting the classification performance of EEG-based pBCI tasks.</p><h2>Other Information</h2><p dir="ltr">Published in: Expert Systems with Applications<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.eswa.2023.120279" target="_blank">https://dx.doi.org/10.1016/j.eswa.2023.120279</a></p>
eu_rights_str_mv openAccess
id Manara2_7a631f74700121872591efb7ef505043
identifier_str_mv 10.1016/j.eswa.2023.120279
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/29445194
publishDate 2023
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling A spectral-ensemble deep random vector functional link network for passive brain–computer interfaceRuilin Li (5627456)Ruobin Gao (16003195)Ponnuthurai N. Suganthan (17347024)Jian Cui (182110)Olga Sourina (17823674)Lipo Wang (17185534)Biomedical and clinical sciencesNeurosciencesEngineeringBiomedical engineeringInformation and computing sciencesArtificial intelligenceEnsemble deep random vector functional link (edRVFL)Spectral-edRVFL (SedRVFL)Electroencephalogram (EEG)Feature-refining (FR) blockDynamic direct link (DDL)<p dir="ltr">Randomized <u>neural networks</u> (RNNs) have shown outstanding performance in many different fields. The superiority of having fewer training parameters and closed-form solutions makes them popular in small datasets analysis. However, automatically decoding raw <u>electroencephalogram</u> (EEG) data using RNNs is still challenging in EEG-based passive brain–computer interface (pBCI) <u>classification tasks</u>. Models with the high-dimension input of EEG may suffer from overfitting and the intrinsic characteristics of non-stationary, high-level noises and subject variability could limit the generation of distinctive features in the hidden layers. To address these problems in EEG-based pBCI tasks, this work proposes a spectral-ensemble deep random vector functional link (SedRVFL) network that focuses on feature learning in the frequency domain. Specifically, an unsupervised feature-refining (FR) block is proposed to improve the low <u>feature learning</u> capability in RNNs. Moreover, a dynamic direct link (DDL) is performed to further complement the <u>frequency information</u>. The proposed model has been evaluated on a self-collected dataset as well as a public driving dataset. The cross-subject <u>classification results</u> obtained demonstrated its effectiveness. This work offers a new solution for EEG decoding, i.e., using optimized RNNs for decoding complex raw EEG data and boosting the classification performance of EEG-based pBCI tasks.</p><h2>Other Information</h2><p dir="ltr">Published in: Expert Systems with Applications<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.eswa.2023.120279" target="_blank">https://dx.doi.org/10.1016/j.eswa.2023.120279</a></p>2023-05-23T15:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.eswa.2023.120279https://figshare.com/articles/journal_contribution/A_spectral-ensemble_deep_random_vector_functional_link_network_for_passive_brain_computer_interface/29445194CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/294451942023-05-23T15:00:00Z
spellingShingle A spectral-ensemble deep random vector functional link network for passive brain–computer interface
Ruilin Li (5627456)
Biomedical and clinical sciences
Neurosciences
Engineering
Biomedical engineering
Information and computing sciences
Artificial intelligence
Ensemble deep random vector functional link (edRVFL)
Spectral-edRVFL (SedRVFL)
Electroencephalogram (EEG)
Feature-refining (FR) block
Dynamic direct link (DDL)
status_str publishedVersion
title A spectral-ensemble deep random vector functional link network for passive brain–computer interface
title_full A spectral-ensemble deep random vector functional link network for passive brain–computer interface
title_fullStr A spectral-ensemble deep random vector functional link network for passive brain–computer interface
title_full_unstemmed A spectral-ensemble deep random vector functional link network for passive brain–computer interface
title_short A spectral-ensemble deep random vector functional link network for passive brain–computer interface
title_sort A spectral-ensemble deep random vector functional link network for passive brain–computer interface
topic Biomedical and clinical sciences
Neurosciences
Engineering
Biomedical engineering
Information and computing sciences
Artificial intelligence
Ensemble deep random vector functional link (edRVFL)
Spectral-edRVFL (SedRVFL)
Electroencephalogram (EEG)
Feature-refining (FR) block
Dynamic direct link (DDL)