|
|
|
 |
|
Liam Paninski
Department of Statistics, Columbia University, USA
Friday 27 June 2008
16.00
Seminar Room B10 (Basement)
Alexandra House, 17 Queen Square, London, WC1N 3AR
A new look at state-space models for neural data
State-space techniques for neural data analysis have become quite popular: these methods are flexible, computationally efficient, and often quite natural from a biophysical point of view. One technical detail is that inference in these models is typically approximate. We introduce some new (faster and more accurate) methods for performing inference in these models and describe a number of applications. Finally, we discuss an extension of these ideas from the one-dimensional temporal domain to some two-dimensional spatial models.