Learning Linear, Sparse, Factorial Codes
dc.date.accessioned | 2004-10-20T20:49:08Z | |
dc.date.accessioned | 2018-11-24T10:23:15Z | |
dc.date.available | 2004-10-20T20:49:08Z | |
dc.date.available | 2018-11-24T10:23:15Z | |
dc.date.issued | 1996-12-01 | en_US |
dc.identifier.uri | http://hdl.handle.net/1721.1/7184 | |
dc.identifier.uri | http://repository.aust.edu.ng/xmlui/handle/1721.1/7184 | |
dc.description.abstract | 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. | en_US |
dc.format.extent | 5 p. | en_US |
dc.format.extent | 233466 bytes | |
dc.format.extent | 268006 bytes | |
dc.language.iso | en_US | |
dc.subject | unsupervised learning | en_US |
dc.subject | factorial coding | en_US |
dc.subject | sparse coding | en_US |
dc.subject | MIT | en_US |
dc.title | Learning Linear, Sparse, Factorial Codes | en_US |
Files in this item
Files | Size | Format | View |
---|---|---|---|
AIM-1580.pdf | 268.0Kb | application/pdf | View/ |
AIM-1580.ps | 233.4Kb | application/postscript | View/ |