Learning Linear, Sparse, Factorial Codes

dc.creator	Olshausen, Bruno A.
dc.date	2004-10-20T20:49:08Z
dc.date	2004-10-20T20:49:08Z
dc.date	1996-12-01
dc.date.accessioned	2013-10-09T02:48:29Z
dc.date.available	2013-10-09T02:48:29Z
dc.date.issued	2013-10-09
dc.identifier	AIM-1580
dc.identifier	CBCL-138
dc.identifier	http://hdl.handle.net/1721.1/7184
dc.identifier.uri	http://koha.mediu.edu.my:8181/xmlui/handle/1721
dc.description	In previous work (Olshausen & Field 1996), an algorithm was described for learning linear sparse codes which, when trained on natural images, produces a set of basis functions that are spatially localized, oriented, and bandpass (i.e., wavelet-like). This note shows how the algorithm may be interpreted within a maximum-likelihood framework. Several useful insights emerge from this connection: it makes explicit the relation to statistical independence (i.e., factorial coding), it shows a formal relationship to the algorithm of Bell and Sejnowski (1995), and it suggests how to adapt parameters that were previously fixed.
dc.format	5 p.
dc.format	233466 bytes
dc.format	268006 bytes
dc.format	application/postscript
dc.format	application/pdf
dc.language	en_US
dc.relation	AIM-1580
dc.relation	CBCL-138
dc.subject	unsupervised learning
dc.subject	factorial coding
dc.subject	sparse coding
dc.subject	MIT
dc.title	Learning Linear, Sparse, Factorial Codes

Files in this item

Files	Size	Format	View
There are no files associated with this item.