Information: Join today and make your research activities more affordable! Technical workshop participation fees and annual registration fees are available at member rates.
Notice: [Important] Announcement of Changes to Registration Fee Payment and Manuscript Upload Procedures for IEICE Technical Meetings
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 2020-12-18 15:10
A Hybrid Sampling Strategy for Improving the Accuracy of Image Classification with less Data
Ruiyun Zhu, Fumihiko Ino (Osaka Univ.) PRMU2020-62
Abstract (in Japanese) (See Japanese page) 
(in English) This paper proposes a hybrid sampling strategy to improve learning accuracy with less training data for image classification tasks.
The hybrid strategy selects valuable samples from unlabeled data set by measuring the uncertainty and diversity of samples so that a high learning accuracy can be reached without traversing the whole data set.
The proposed method measures the amount of information of each sample by using the autoencoder technique and Wasserstein metric, which estimate the uncertainty and diversity of samples, respectively.
More specifically, we propose a neural network that consists of four components: a feature extractor, a Wasserstein distance estimator, an image classifier, and an image reconstructor.
These components are responsible for feature extraction, distribution matching, image classification, and image reconstruction, respectively.
Experimental results with the Fashion MNIST data set indicate that the proposed method improves the learning accuracy from 74.7% to 77.4% compared with a random sampling method.
With a relatively complicated data set, CIFAR-10, the learning accuracy of the proposed method increases from 65.2% to 68.4% compared to that of the previous method.
Keyword (in Japanese) (See Japanese page) 
(in English) Image classification / Autoencoder / Wasserstein metric / / / / /  
Reference Info. IEICE Tech. Rep., vol. 120, no. 300, PRMU2020-62, pp. 139-144, Dec. 2020.
Paper # PRMU2020-62 
Date of Issue 2020-12-10 (PRMU) 
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 PRMU2020-62

Conference Information
Committee PRMU  
Conference Date 2020-12-17 - 2020-12-18 
Place (in Japanese) (See Japanese page) 
Place (in English) Online 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Transfer learning and few shot learning 
Paper Information
Registration To PRMU 
Conference Code 2020-12-PRMU 
Language English (Japanese title is available) 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A Hybrid Sampling Strategy for Improving the Accuracy of Image Classification with less Data 
Sub Title (in English)  
Keyword(1) Image classification  
Keyword(2) Autoencoder  
Keyword(3) Wasserstein metric  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Ruiyun Zhu  
1st Author's Affiliation Osaka University (Osaka Univ.)
2nd Author's Name Fumihiko Ino  
2nd Author's Affiliation Osaka University (Osaka Univ.)
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 ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2020-12-18 15:10:00 
Presentation Time 15 minutes 
Registration for PRMU 
Paper # PRMU2020-62 
Volume (vol) vol.120 
Number (no) no.300 
Page pp.139-144 
#Pages 6 
Date of Issue 2020-12-10 (PRMU) 

[Return to Top Page]

[Return to IEICE Web Page]
The Institute of Electronics, Information and Communication Engineers (IEICE), Japan