Presentation 2006-02-24
Fast Learning Approach Using Self Organizing Map
Alaa SAGHEER, Nayouki TSURUTA, Rin-Ichiro TANIGUCHI, Daisaku ARITA, Sakashi MAEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The Self Organizing Map (SOM) is one of the most widely used neural network paradigm based on unsupervised competitive learning. However, the learning algorithm introduced by Kohonen is very slow when the size of the map is large. This slowness is caused by seeking about the best node among "all" the map nodes which tunes to "each" input sample. In this paper, a novel fast learning SOM algorithm is proposed. Exploiting a new strategy, the new algorithm runs by concerning "only" about the nodes which are aligned around principal components and neglects the rest of nodes which already include less information [1]. Experimental results are reported at the end of this paper. Two data sets are utilized to illustrate the proposed algorithm. Under same experiment conditions, it is shown here that the computation time is reduced to O(log N) instead of O(N). Also our method computation time is less than that of FDCT by 6 times under same experimental conditions.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Self Organizing Map / Lip-Reading Systems / Principal Components Theory
Paper # TL2005-71,PRMU2005-206
Date of Issue

Conference Information
Committee TL
Conference Date 2006/2/17(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 Thought and Language (TL)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Fast Learning Approach Using Self Organizing Map
Sub Title (in English)
Keyword(1) Self Organizing Map
Keyword(2) Lip-Reading Systems
Keyword(3) Principal Components Theory
1st Author's Name Alaa SAGHEER
1st Author's Affiliation Department of Intelligent Systems, Kyushu University()
2nd Author's Name Nayouki TSURUTA
2nd Author's Affiliation Department of Electronics Engineering and Computer Science, Fukuoka University
3rd Author's Name Rin-Ichiro TANIGUCHI
3rd Author's Affiliation Department of Intelligent Systems, Kyushu University
4th Author's Name Daisaku ARITA
4th Author's Affiliation Department of Intelligent Systems, Kyushu University
5th Author's Name Sakashi MAEDA
5th Author's Affiliation Department of Electronics Engineering and Computer Science, Fukuoka University
Date 2006-02-24
Paper # TL2005-71,PRMU2005-206
Volume (vol) vol.105
Number (no) 613
Page pp.pp.-
#Pages 6
Date of Issue