Ghent University Academic Bibliography

Advanced

A P300 BCI for the masses: prior information enables instant unsupervised spelling

Pieter-Jan Kindermans UGent, Hannes Verschore UGent, David Verstraeten UGent and Benjamin Schrauwen UGent (2012) Advances in Neural Information Processing Systems 25.
abstract
The usability of Brain Computer Interfaces (BCI) based on the P300 speller is severely hindered by the need for long training times and many repetitions of the same stimulus. In this contribution we introduce a set of unsupervised hierarchical probabilistic models that tackle both problems simultaneously by incorporating prior knowledge from two sources: information from other training subjects (through transfer learning) and information about the words being spelled (through language models). We show, that due to this prior knowledge, the performance of the unsupervised models parallels and in some cases even surpasses that of supervised models, while eliminating the tedious training session.
Please use this url to cite or link to this publication:
author
organization
year
type
conference
publication status
in press
subject
keyword
Unsupervised, BCI, Brain Computer Interface, Machine Learning, graphical model, Bayesian model, Regression
in
Advances in Neural Information Processing Systems 25
pages
9 pages
publisher
Ghent University, Department of Electronics and information systems
place of publication
Ghent, Belgium
conference name
Neural Information Processing Systems (NIPS - 2012)
conference location
Lake Tahoe, Nevada, USA
conference start
2012-12-03
conference end
2012-12-06
language
English
UGent publication?
yes
classification
C1
copyright statement
I have transferred the copyright for this publication to the publisher
id
3008919
handle
http://hdl.handle.net/1854/LU-3008919
date created
2012-10-08 15:43:27
date last changed
2012-10-15 13:08:57
@inproceedings{3008919,
  abstract     = {The usability of Brain Computer Interfaces (BCI) based on the P300 speller is severely hindered by the need for long training times and many repetitions of the same stimulus. In this contribution we introduce a set of unsupervised hierarchical probabilistic models that tackle both problems simultaneously by incorporating prior knowledge from two sources: information from other training subjects (through transfer learning) and information about the words being spelled (through language models). We show, that due to this prior knowledge, the performance of the unsupervised models parallels and in some cases even surpasses that of supervised models, while eliminating the tedious training session.},
  author       = {Kindermans, Pieter-Jan and Verschore, Hannes and Verstraeten, David and Schrauwen, Benjamin},
  booktitle    = {Advances in Neural Information Processing Systems 25},
  keyword      = {Unsupervised,BCI,Brain Computer Interface,Machine Learning,graphical model,Bayesian model,Regression},
  language     = {eng},
  location     = {Lake Tahoe, Nevada, USA},
  pages        = {9},
  publisher    = {Ghent University, Department of Electronics and information systems},
  title        = {A P300 BCI for the masses: prior information enables instant unsupervised spelling},
  year         = {2012},
}

Chicago
Kindermans, Pieter-Jan, Hannes Verschore, David Verstraeten, and Benjamin Schrauwen. 2012. “A P300 BCI for the Masses: Prior Information Enables Instant Unsupervised Spelling.” In Advances in Neural Information Processing Systems 25. Ghent, Belgium: Ghent University, Department of Electronics and information systems.
APA
Kindermans, P.-J., Verschore, H., Verstraeten, D., & Schrauwen, B. (2012). A P300 BCI for the masses: prior information enables instant unsupervised spelling. Advances in Neural Information Processing Systems 25. Presented at the Neural Information Processing Systems (NIPS - 2012), Ghent, Belgium: Ghent University, Department of Electronics and information systems.
Vancouver
1.
Kindermans P-J, Verschore H, Verstraeten D, Schrauwen B. A P300 BCI for the masses: prior information enables instant unsupervised spelling. Advances in Neural Information Processing Systems 25. Ghent, Belgium: Ghent University, Department of Electronics and information systems; 2012.
MLA
Kindermans, Pieter-Jan, Hannes Verschore, David Verstraeten, et al. “A P300 BCI for the Masses: Prior Information Enables Instant Unsupervised Spelling.” Advances in Neural Information Processing Systems 25. Ghent, Belgium: Ghent University, Department of Electronics and information systems, 2012. Print.