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 2026-03-02 14:00
[Poster Presentation] A study on a k-medoids-based sampling strategy for efficient training of auditory perception models -- An evaluation on reverberation time prediction --
Takumi Koga, Natsuki Ueno (Kumamoto Univ.), Kenji Ishizuka, Akito Nakamura, Yu Takahashi (Yamaha Corp.) EA2025-87 SIP2025-107 SP2025-40
Abstract (in Japanese) (See Japanese page) 
(in English) When constructing an auditory perceptual model that emulates human subjective judgments using machine learning, pairwise-comparison-based annotation, where two data samples are compared and only their relative preference is judged, is effective because it does not require an absolute rating scale. However, when the number of candidate pairs to be annotated is enormous, the burden on participants becomes substantial; therefore, it is crucial to appropriately select informative pairs that enable efficient learning.
In this study, we propose a sampling strategy based on k-medoids clustering. The proposed method aims to reduce the number of comparison pairs while maintaining data diversity, thereby enabling efficient learning of the global ordering among all samples even under a limited number of comparisons. Furthermore, through an experiment on predicting the reverberation time $mathrm{RT}_{60}$ from room impulse responses, we confirmed that the proposed method achieves higher predictive performance than random sampling and improves learning efficiency under the same constraint on the number of comparisons.
Keyword (in Japanese) (See Japanese page) 
(in English) k-medoids / RankNet / learning-to-rank / auditory perceptual model / / / /  
Reference Info. IEICE Tech. Rep., vol. 125, no. 369, EA2025-87, pp. 87-92, March 2026.
Paper # EA2025-87 
Date of Issue 2026-02-23 (EA, SIP, SP) 
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 EA2025-87 SIP2025-107 SP2025-40

Conference Information
Committee SP EA SIP IPSJ-SLP  
Conference Date 2026-03-02 - 2026-03-04 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To EA 
Conference Code 2026-03-SP-EA-SIP-SLP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A study on a k-medoids-based sampling strategy for efficient training of auditory perception models 
Sub Title (in English) An evaluation on reverberation time prediction 
Keyword(1) k-medoids  
Keyword(2) RankNet  
Keyword(3) learning-to-rank  
Keyword(4) auditory perceptual model  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Takumi Koga  
1st Author's Affiliation Kumamoto University (Kumamoto Univ.)
2nd Author's Name Natsuki Ueno  
2nd Author's Affiliation Kumamoto University (Kumamoto Univ.)
3rd Author's Name Kenji Ishizuka  
3rd Author's Affiliation Yamaha Corporation (Yamaha Corp.)
4th Author's Name Akito Nakamura  
4th Author's Affiliation Yamaha Corporation (Yamaha Corp.)
5th Author's Name Yu Takahashi  
5th Author's Affiliation Yamaha Corporation (Yamaha Corp.)
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 2026-03-02 14:00:00 
Presentation Time 80 minutes 
Registration for EA 
Paper # EA2025-87, SIP2025-107, SP2025-40 
Volume (vol) vol.125 
Number (no) no.369(EA), no.370(SIP), no.371(SP) 
Page pp.87-92 
#Pages 6 
Date of Issue 2026-02-23 (EA, SIP, SP) 

[Return to Top Page]

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