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 2021-03-02 10:00
Learning from Noisy Complementary Labels with Robust Loss Functions
Hiroki Ishiguro (UTokyo), Takashi Ishida (UTokyo/RIKEN), Masashi Sugiyama (RIKEN/UTokyo) IBISML2020-34
Abstract (in Japanese) (See Japanese page) 
(in English) It has been demonstrated that large-scale labeled datasets facilitate the success of machine learning. However, collecting labeled data is often very costly and error-prone in practice. To cope with this problem, previous studies have considered the use of a complementary label, which specifies a class that an instance does not belong to and can be collected more easily than ordinary labels. However, complementary labels could also be error-prone and thus mitigating the influence of label noise is an important challenge to make complementary-label learning more useful in practice. In this paper, we derive conditions for the loss function such that the learning algorithm is not affected by noise in complementary labels. Experiments on benchmark datasets with noisy complementary labels demonstrate that the loss functions that satisfy our conditions significantly improve the classification performance.
Keyword (in Japanese) (See Japanese page) 
(in English) complementary label / label noise / robust loss function / loss correction / / / /  
Reference Info. IEICE Tech. Rep., vol. 120, no. 395, IBISML2020-34, pp. 1-8, March 2021.
Paper # IBISML2020-34 
Date of Issue 2021-02-23 (IBISML) 
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 IBISML2020-34

Conference Information
Committee IBISML  
Conference Date 2021-03-02 - 2021-03-04 
Place (in Japanese) (See Japanese page) 
Place (in English) Online 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Organized and general sessions on machine learning 
Paper Information
Registration To IBISML 
Conference Code 2021-03-IBISML 
Language English (Japanese title is available) 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Learning from Noisy Complementary Labels with Robust Loss Functions 
Sub Title (in English)  
Keyword(1) complementary label  
Keyword(2) label noise  
Keyword(3) robust loss function  
Keyword(4) loss correction  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Hiroki Ishiguro  
1st Author's Affiliation The University of Tokyo (UTokyo)
2nd Author's Name Takashi Ishida  
2nd Author's Affiliation The University of Tokyo/RIKEN (UTokyo/RIKEN)
3rd Author's Name Masashi Sugiyama  
3rd Author's Affiliation RIKEN/The University of Tokyo (RIKEN/UTokyo)
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 2021-03-02 10:00:00 
Presentation Time 25 minutes 
Registration for IBISML 
Paper # IBISML2020-34 
Volume (vol) vol.120 
Number (no) no.395 
Page pp.1-8 
#Pages
Date of Issue 2021-02-23 (IBISML) 


[Return to Top Page]

[Return to IEICE Web Page]


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