Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-6601
Main Title: Learning from label proportions in brain-computer interfaces
Subtitle: online unsupervised learning with guarantees
Author(s): Hübner, David
Verhoeven, Thibault
Schmid, Konstantin
Müller, Klaus-Robert
Tangermann, Michael
Kindermans, Pieter-Jan
Type: Article
Language Code: en
Abstract: Objective Using traditional approaches, a brain-computer interface (BCI) requires the collection of calibration data for new subjects prior to online use. Calibration time can be reduced or eliminated e.g., by subject-to-subject transfer of a pre-trained classifier or unsupervised adaptive classification methods which learn from scratch and adapt over time. While such heuristics work well in practice, none of them can provide theoretical guarantees. Our objective is to modify an event-related potential (ERP) paradigm to work in unison with the machine learning decoder, and thus to achieve a reliable unsupervised calibrationless decoding with a guarantee to recover the true class means. Method We introduce learning from label proportions (LLP) to the BCI community as a new unsupervised, and easy-to-implement classification approach for ERP-based BCIs. The LLP estimates the mean target and non-target responses based on known proportions of these two classes in different groups of the data. We present a visual ERP speller to meet the requirements of LLP. For evaluation, we ran simulations on artificially created data sets and conducted an online BCI study with 13 subjects performing a copy-spelling task. Results Theoretical considerations show that LLP is guaranteed to minimize the loss function similar to a corresponding supervised classifier. LLP performed well in simulations and in the online application, where 84.5% of characters were spelled correctly on average without prior calibration. Significance The continuously adapting LLP classifier is the first unsupervised decoder for ERP BCIs guaranteed to find the optimal decoder. This makes it an ideal solution to avoid tedious calibration sessions. Additionally, LLP works on complementary principles compared to existing unsupervised methods, opening the door for their further enhancement when combined with LLP.
URI: https://depositonce.tu-berlin.de//handle/11303/7328
http://dx.doi.org/10.14279/depositonce-6601
Issue Date: 2017
Date Available: 17-Jan-2018
DDC Class: 006 Spezielle Computerverfahren
Subject(s): event-related potentials
electroencephalography
man-computer interface
algorithms
machine learning algorithms
covariance
machine learning
preprocessing
Sponsor/Funder: DFG, TH 662/19-1, Open Access Publizieren 2017 - 2018 / Technische Universität Berlin
License: https://creativecommons.org/licenses/by/4.0/
Journal Title: PLoS one
Publisher: PLoS
Publisher Place: Lawrence, Kan.
Volume: 12
Issue: 4
Article Number: e0175856
Publisher DOI: 10.1371/journal.pone.0175856
ISSN: 1932-6203
Appears in Collections:FG Maschinelles Lernen » Publications

Files in This Item:
File Description SizeFormat 
2017_kindermans_et-al.pdf3.68 MBAdobe PDFThumbnail
View/Open


This item is licensed under a Creative Commons License Creative Commons