IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2010-03-11 17:15
Analysis of Autocorrelation type Associative Memory Model with Hierarchical Patterns by Using PCA
Teijiro Shiotsuka (Waseda Univ.), Kenji Nagata, Koji Hukushima (Tokyo Univ.), Masato Okada (Tokyo Univ./RIKEN), Masato Inoue (Waseda Univ.) NC2009-158
Abstract (in Japanese) (See Japanese page) 
(in English) The statistical mechanical approach is useful and has been applied to various problems in the field of information processing, but its application is mainly limited to the class of the mean-field models. This report proposes a more general analysis method and validates its effectiveness. This method combines exchange Monte Carlo method and principal component analysis (PCA), and visualizes obtained high-dimensional empirical distribution. Through this method, we try to estimate the macro state of a given system. We also estimate an effective number of principal components by introducing probabilistic principal component analysis (PPCA). We have applied this approach to associative memory model, and obtained consistent results with those by the mean-field analysis. Those results suggest the possibility that we can analyze not-analytically-solved systems by using this approach.
Keyword (in Japanese) (See Japanese page) 
(in English) statistical mechanics / exchange Monte Carlo method / principal component analysis / probabilistic principal component analysis / autocorrelation type associative memory model / hierarchical patterns / /  
Reference Info. IEICE Tech. Rep., vol. 109, no. 461, NC2009-158, pp. 413-418, March 2010.
Paper # NC2009-158 
Date of Issue 2010-03-02 (NC) 
ISSN Print edition: ISSN 0913-5685    Online edition: ISSN 2432-6380
Copyright
and
reproduction
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF NC2009-158

Conference Information
Committee NC MBE  
Conference Date 2010-03-09 - 2010-03-11 
Place (in Japanese) (See Japanese page) 
Place (in English) Tamagawa University 
Topics (in Japanese) (See Japanese page) 
Topics (in English) General 
Paper Information
Registration To NC 
Conference Code 2010-03-NC-MBE 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Analysis of Autocorrelation type Associative Memory Model with Hierarchical Patterns by Using PCA 
Sub Title (in English)  
Keyword(1) statistical mechanics  
Keyword(2) exchange Monte Carlo method  
Keyword(3) principal component analysis  
Keyword(4) probabilistic principal component analysis  
Keyword(5) autocorrelation type associative memory model  
Keyword(6) hierarchical patterns  
Keyword(7)  
Keyword(8)  
1st Author's Name Teijiro Shiotsuka  
1st Author's Affiliation Waseda University (Waseda Univ.)
2nd Author's Name Kenji Nagata  
2nd Author's Affiliation Tokyo University (Tokyo Univ.)
3rd Author's Name Koji Hukushima  
3rd Author's Affiliation Tokyo University (Tokyo Univ.)
4th Author's Name Masato Okada  
4th Author's Affiliation Tokyo University/RIKEN Brain Science Institute (Tokyo Univ./RIKEN)
5th Author's Name Masato Inoue  
5th Author's Affiliation Waseda University (Waseda Univ.)
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2010-03-11 17:15:00 
Presentation Time 25 minutes 
Registration for NC 
Paper # NC2009-158 
Volume (vol) vol.109 
Number (no) no.461 
Page pp.413-418 
#Pages
Date of Issue 2010-03-02 (NC) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan