Neural coding information sought

Mark Laubach laubach at biogfx.neuro.wfu.edu
Fri Dec 8 15:45:53 EST 1995


laubach at biogfx.neuro.wfu.edu (Mark Laubach) wrote:

>Nonlinear signal processing methods
>developed by researchers in the wavelets community, such as the best
>basis method, matching pursuit, local discriminant bases, and
>discriminant pursuit, allow one to assess the _time-frequency
>structure_ of signals such as neural spike trains.  The only choice to
>make is how to filter (i.e., smooth) the spike train prior to these
>analyses.  

I was not entirely correct here.  Wavelet methods also require the
analyst to make choices about certain parameters.  However, the best
parameters for the analysis can be determined empirically.  For
example, the choice of parameters for local discriminant bases, e.g.,
mother wavelet, entropy criterion, etc. can be optimized by comparing
the classification success of the coefficients (features) chosen by
the algorithm.
-----------------------------------
Mark Laubach
Dept. of Physiology & Pharmacology
Bowman Gray School of Medicine
Wake Forest University
Winston-Salem, NC 27157
laubach at biogfx.neuro.wfu.edu
-----------------------------------




More information about the Neur-sci mailing list