Presentation 2004/6/17
Independent Component Analysis and Principal Component Analysis by a Coupled Pair of One-dimensional Self-Organizing Maps
Takashi TAJIRI, Koji KURATA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Shirakura and Kurata proposed a model for information separation consisting of two self-organizing maps, which are one-dimensional each and connected by plastic inhibitory connections. With inputs parametrized by two parameters the model could form a map of each parameter on each one-dimensional self-organizing map. Slightly modified for better performance but essentially a same model was examined in detail to clarify what kind of information separation is achieved by the model. Two versions of the model was numerically studied and compared. Feedforward inhibition model was shown to organize maps of PCA-like representation, while recurrent inhibition model organized ICA-like representation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Self-Organizing Maps / Nerve field / Independent Component Analysis / Principal Component Analysis
Paper # NC2004-22
Date of Issue

Conference Information
Committee NC
Conference Date 2004/6/17(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Independent Component Analysis and Principal Component Analysis by a Coupled Pair of One-dimensional Self-Organizing Maps
Sub Title (in English)
Keyword(1) Self-Organizing Maps
Keyword(2) Nerve field
Keyword(3) Independent Component Analysis
Keyword(4) Principal Component Analysis
1st Author's Name Takashi TAJIRI
1st Author's Affiliation Graduate school of Engineering and Science, University of the Ryukyus()
2nd Author's Name Koji KURATA
2nd Author's Affiliation Faculty of Engineering, University of the Ryukyus
Date 2004/6/17
Paper # NC2004-22
Volume (vol) vol.104
Number (no) 139
Page pp.pp.-
#Pages 6
Date of Issue