Taira NAKAJIMA


Vector Quantization Codebook Design Using the Law-of-the-Jungle Algorithm
Hiroyuki TAKIZAWA Taira NAKAJIMA Kentaro SANO Hiroaki KOBAYASHI Tadao NAKAMURA 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 2003/06/01
Vol. E86-D  No. 6  pp. 1068-1077
Type of Manuscript:  PAPER
Category: Image Processing, Image Pattern Recognition
Keyword: 
vector quantizationthe equidistortion principlethe law-of-the-jungle algorithmthe minimax partial distortion criterion
 Summary | Full Text:PDF(557.2KB)

An Active Learning Algorithm Based on Existing Training Data
Hiroyuki TAKIZAWA Taira NAKAJIMA Hiroaki KOBAYASHI Tadao NAKAMURA 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 2000/01/25
Vol. E83-D  No. 1  pp. 90-99
Type of Manuscript:  PAPER
Category: Biocybernetics, Neurocomputing
Keyword: 
multilayer perceptronsthe network inversion algorithmactive learningclassification problems
 Summary | Full Text:PDF(570.4KB)

A Topology Preserving Neural Network for Nonstationary Distributions
Taira NAKAJIMA Hiroyuki TAKIZAWA Hiroaki KOBAYASHI Tadao NAKAMURA 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 1999/07/25
Vol. E82-D  No. 7  pp. 1131-1135
Type of Manuscript:  LETTER
Category: Bio-Cybernetics and Neurocomputing
Keyword: 
competitive Hebbian learning rulelaw of the jungle mechanismneural networknonstationary probability distributionself-organizing map
 Summary | Full Text:PDF(209.4KB)

Acceleration Techniques for the Network Inversion Algorithm
Hiroyuki TAKIZAWA Taira NAKAJIMA Masaaki NISHI Hiroaki KOBAYASHI Tadao NAKAMURA 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 1999/02/25
Vol. E82-D  No. 2  pp. 508-511
Type of Manuscript:  LETTER
Category: Bio-Cybernetics and Neurocomputing
Keyword: 
multilayer perceptronnetwork inversion algorithmactive learning based on existing training examplesclassification problems
 Summary | Full Text:PDF(133.3KB)

Kohonen Learning with a Mechanism, the Law of the Jungle, Capable of Dealing with Nonstationary Probability Distribution Functions
Taira NAKAJIMA Hiroyuki TAKIZAWA Hiroaki KOBAYASHI Tadao NAKAMURA 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 1998/06/25
Vol. E81-D  No. 6  pp. 584-591
Type of Manuscript:  PAPER
Category: Bio-Cybernetics and Neurocomputing
Keyword: 
adaptive vector quantizationequiprobable principleconscience mechanism Kohonen learningLOJ mechanismneural networksnonstationary probability distribution
 Summary | Full Text:PDF(617.7KB)