Learning from label proportions in brain-computer interfaces: Online unsupervised learning with guarantees.

Journal: PloS one
Published Date:

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.

Authors

  • David Hübner
    Brain State Decoding Lab, Cluster of Excellence BrainLinks-BrainTools, Department of Computer Science, Albert-Ludwigs-University, Freiburg, Germany.
  • Thibault Verhoeven
    Department of Electronics and Information Systems, Ghent University, Ghent, Belgium. Electronic address: thibault.verhoeven@ugent.be.
  • Konstantin Schmid
    Brain State Decoding Lab, Cluster of Excellence BrainLinks-BrainTools, Department of Computer Science, Albert-Ludwigs-University, Freiburg, Germany.
  • Klaus-Robert Müller
    Berlin Institute for the Foundations of Learning and Data (BIFOLD), Berlin, Deutschland.
  • Michael Tangermann
    Brain State Decoding Lab, Cluster of Excellence BrainLinks-BrainTools, Department of Computer Science, Albert-Ludwigs-University, Freiburg, Germany.
  • Pieter-Jan Kindermans