BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Margin- and Evidence-Based Approaches for EEG Signal Classificatio
 n - N. Jeremy Hill\, Max Planck Institute for Biological Cybernetics
DTSTART:20070209T150000Z
DTEND:20070209T160000Z
UID:TALK6672@talks.cam.ac.uk
CONTACT:Oliver Williams
DESCRIPTION:Our principal motivation in Brain-Computer Interface (BCI) res
 earch is to develop systems that will enable a completely paralysed person
  to communicate. Many projects have adopted the approach of classifying ch
 anges in the bandpower in certain parts of the spectrum of brain signals m
 easured by electroencephalogram (EEG). Appropriate preprocessing\, in the 
 form of spatial and temporal filtering\, is crucial for this\, and general
 ly the `conventional wisdom` of the BCI community is (a) that once the pre
 processing is right\, it doesn`t matter which classifier you use and accor
 dingly (b) that a linear classifier will generally perform as well as any 
 non-linear one. The hitherto most successful preprocessing algorithm\, the
  Common Spatial Pattern algorithm\, is a supervised method for computing s
 patial filters very cheaply. However\, it uses a least-square criterion an
 d is very prone to overfitting with small amounts of data. Our approach is
  to combine feature extraction and classification into a single optimizati
 on step\, and to optimize a criterion that is a better predictor of genera
 lization performance: namely the margin (as in the Support Vector Machine)
  or the evidence (a.k.a. marginal likelihood\, in this case obtained from 
 a Gaussian Process classifier). I will show that this yields consistent im
 provements in performance\, particularly in the (most clinically relevant)
  cases where data are noisy and/or few in number\, and that projection int
 o a higher-dimensional feature space via a non-linear kernel can improve p
 erformance further. I will also show a preliminary demonstration that the 
 approach can simultaneously recover optimal weightings across space\, freq
 uency and time\, with little sensitivity to prior assumptions: this makes 
 it a promising tool for the analysis of biosignal data in general. \n\n
LOCATION:Large public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR