Friday, December 10, 2004

12 noon

Redwood Neuroscience Institute

 

Title "Is slowness a learning principle of visual cortex?"

 

Laurenz Wiskott

Institute for Theoretical Biology

Humboldt University Berlin

 

Abstract:

Slow Feature Analysis (SFA) is an algorithm for extracting slowly varying features from a quickly varying signal.  It has been shown in network simulations on 1-dimensional stimuli that visual invariances to shift, scaling, illumination and other transformations can be learned in an unsupervised fashion based on SFA [1].

 

More recently we have shown that SFA applied to image sequences generated from natural images using a range of spatial transformations results in units that share many properties of complex and hypercomplex cells of early visual areas [2].  We find cells responsive to Gabor stimuli with phase invariance, sharpened or widened orientation or frequency tuning, secondary response lobes, end-stopping, and cells selective for direction of motion.

 

These results indicate that slowness may be an important principle of self-organization in the visual cortex.

 

 

[1] Wiskott, L. and Sejnowski, T.J. (2002).  Slow Feature Analysis: Unsupervised Learning of Invariances.  Neural Computation, 14(4):715-770. http://itb1.biologie.hu-berlin.de/~wiskott/Abstracts/WisSej2002.html

 

[2] Berkes, P. and Wiskott, L. (2003).  Slow feature analysis yields a rich repertoire of complex-cell properties.  Cognitive Sciences EPrint Archive (CogPrints) 2804, http://cogprints.ecs.soton.ac.uk/archive/00002804/

http://itb1.biologie.hu-berlin.de/~wiskott/Abstracts/BerkWisk2003a.html