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 2019-03-06 15:00
Efficient Learning for Distillation of DNN by Self Distillation
Jumpei Takagi, Motonobu Hattori (Univ of Yamanashi) NC2018-83
Abstract (in Japanese) (See Japanese page) 
(in English) Knowledge distillation is a method to create a superior student by using knowledge obtained from a trained teacher neural network. Recent studies have shown that much superior students can be obtained by distilling the trained student further as a teacher. Distilling the knowledge through multiple generations, however, takes a long time for learning. In this paper, we propose a self distillation method which can reduce both the number of generations and learning time for knowledge distillation. In self distillation, the most accurate network is obtained during intra-generation learning, and it is used as a teacher of intra-generational distillation. Our experiments for image classification task demonstrate that the proposed self distillation acquires high accuracy with fewer generations and less learning time than the conventional method.
Keyword (in Japanese) (See Japanese page) 
(in English) Knowledge Distillation / Self Distillation / Deep Learning / Image Classification / / / /  
Reference Info. IEICE Tech. Rep., vol. 118, no. 470, NC2018-83, pp. 209-214, March 2019.
Paper # NC2018-83 
Date of Issue 2019-02-25 (NC) 
ISSN Print edition: ISSN 0913-5685  Online edition: ISSN 2432-6380
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 NC2018-83

Conference Information
Committee NC MBE  
Conference Date 2019-03-04 - 2019-03-06 
Place (in Japanese) (See Japanese page) 
Place (in English) University of Electro Communications 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To NC 
Conference Code 2019-03-NC-MBE 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Efficient Learning for Distillation of DNN by Self Distillation 
Sub Title (in English)  
Keyword(1) Knowledge Distillation  
Keyword(2) Self Distillation  
Keyword(3) Deep Learning  
Keyword(4) Image Classification  
1st Author's Name Jumpei Takagi  
1st Author's Affiliation University of Yamanashi (Univ of Yamanashi)
2nd Author's Name Motonobu Hattori  
2nd Author's Affiliation University of Yamanashi (Univ of Yamanashi)
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 ()
Date Time 2019-03-06 15:00:00 
Presentation Time 25 
Registration for NC 
Paper # IEICE-NC2018-83 
Volume (vol) IEICE-118 
Number (no) no.470 
Page pp.209-214 
#Pages IEICE-6 
Date of Issue IEICE-NC-2019-02-25 

[Return to Top Page]

[Return to IEICE Web Page]

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