DocumentCode :
2136697
Title :
A symmetry-breaking generative model of a simple-cell/complex-cell hierarchy
Author :
Schultz, Peter F. ; Bettencourt, Luis M. ; Kenyon, Garrett T.
Author_Institution :
New Mexico Consortium, Los Alamos, NM, USA
fYear :
2012
fDate :
22-24 April 2012
Firstpage :
89
Lastpage :
92
Abstract :
We present a generative model for a three-layer hierarchy consisting of a retinal layer, a simple-cell layer, and a complex-cell layer. The weights in the model are trained using supervised learning on the retinal layer and complex cells. Once the weights are learned, the model is able to perform bottom-up classification of images, as well as top-down reconstruction from a specified category. The symmetry-breaking aspect of the model prevents the top-down reconstruction from generating images with a mixture of incompatible features. We illustrate the performance of this model with an image space consisting of vertical and horizontal bars in varying positions, in which the complex-cell layer learns the invariance that groups all horizontal bars into one category and all vertical into another.
Keywords :
image classification; image reconstruction; learning (artificial intelligence); bottom-up image classification; complex-cell hierarchy; complex-cell layer; image space; retinal layer; simple-cell hierarchy; simple-cell layer; supervised learning; symmetry-breaking generative model; three-layer hierarchy; top-down reconstruction; Adaptation models; Bars; Convergence; Image reconstruction; Neurons; Training; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Analysis and Interpretation (SSIAI), 2012 IEEE Southwest Symposium on
Conference_Location :
Santa Fe, NM
Print_ISBN :
978-1-4673-1831-0
Electronic_ISBN :
978-1-4673-1829-7
Type :
conf
DOI :
10.1109/SSIAI.2012.6202460
Filename :
6202460
Link To Document :
بازگشت