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 2019-12-06 15:40
Prevention of redundant representations and of the black box in stacked autoencoders
Masumi Ishikawa (Kyutech) MBE2019-56 NC2019-47
Abstract (in Japanese) (See Japanese page) 
(in English) Recent progress in deep learning (DL) is remarkable and its recognition capability is said to surpass that of humans. The acquired features, however, don’t seem to be sufficiently clear. Furthermore, the basis of judgement by DL is mostly unknown. In 1990s the author proposed L1-norm regularization and relevant terms to create understandable neural networks mainly with discrete valued inputs and outputs. Because DL is inherently nonlinear, L1-norm alone might not be enough to understand the resulting models due to frequently emerging redundant representations. The present paper extends the previous proposal to deep learning models with special emphasis on redundant representations. Training with L1-norm and selective L1-norm to connection weights, decomposition of a model, and suppressing redundant representations are proposed. The paper clarifies the effectiveness of the proposal by applying it to stacked autoencoders using red wine quality data.
Keyword (in Japanese) (See Japanese page) 
(in English) Stacked autoencoder / Sparse modeling / Deep learning / Redundant representation / Black box / / /  
Reference Info. IEICE Tech. Rep., vol. 119, no. 328, NC2019-47, pp. 67-72, Dec. 2019.
Paper # NC2019-47 
Date of Issue 2019-11-29 (MBE, NC) 
ISSN 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 MBE2019-56 NC2019-47

Conference Information
Committee NC MBE  
Conference Date 2019-12-06 - 2019-12-06 
Place (in Japanese) (See Japanese page) 
Place (in English) Toyohashi Tech 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To NC 
Conference Code 2019-12-NC-MBE 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Prevention of redundant representations and of the black box in stacked autoencoders 
Sub Title (in English)  
Keyword(1) Stacked autoencoder  
Keyword(2) Sparse modeling  
Keyword(3) Deep learning  
Keyword(4) Redundant representation  
Keyword(5) Black box  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Masumi Ishikawa  
1st Author's Affiliation Kyushu Institute of Technology (Kyutech)
2nd Author's Name  
2nd Author's Affiliation ()
3rd Author's Name  
3rd Author's Affiliation ()
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
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 2019-12-06 15:40:00 
Presentation Time 25 minutes 
Registration for NC 
Paper # MBE2019-56, NC2019-47 
Volume (vol) vol.119 
Number (no) no.327(MBE), no.328(NC) 
Page pp.67-72 
#Pages
Date of Issue 2019-11-29 (MBE, NC) 


[Return to Top Page]

[Return to IEICE Web Page]


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