Adaptive classification for Brain Computer Interface systems using Sequential Monte Carlo sampling

Ji Won Yoon, Stephen J. Roberts, Matt Dyson, John Q. Gan

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

Adaptive classification is a key function of Brain Computer Interfacing (BCI) systems. This paper proposes robust mathematical frameworks and their implementation for the on-line sequential classification of EEG signals in BCI systems. The proposed algorithms are extensions to the basic method of Andrieu et al. [Andrieu, C., de Freitas, N., and Doucet, A. (2001). Sequential bayesian semi-parametric binary classification. In Proc. NIPS], modified to be suitable for BCI use. We focus on the inference and prediction of target labels under a non-linear and non-Gaussian model. In this paper we introduce two new algorithms to handle missing or erroneous labeling in BCI data. One algorithm introduces auxiliary labels to process the uncertainty of the labels and the other modifies the optimal proposal functions to allow for uncertain labels. Although we focus on BCI problems in this paper, the algorithms can be generalized and applied to other application domains in which sequential missing labels are to be imputed under the presence of uncertainty.

Original languageEnglish
Pages (from-to)1286-1294
Number of pages9
JournalNeural Networks
Volume22
Issue number9
DOIs
Publication statusPublished - 2009 Nov
Externally publishedYes

Keywords

  • Auxiliary labels
  • Dynamic probit model
  • Missing labels
  • Nonlinear dynamics
  • Online Brain Computer Interfacing (BCI)
  • Rao-Blackwellised Sequential Monte Carlo

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Adaptive classification for Brain Computer Interface systems using Sequential Monte Carlo sampling'. Together they form a unique fingerprint.

  • Cite this