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 2026-03-25 09:20
Analysis of the Visual Capabilities of a Deep Neural Network Trained via Reinforcement Learning for Autonomous Behavior in a Virtual Environment
Naoko Kaneda, Sena Sato, Makoto murakami (Toyo univ) PRMU2025-56
Abstract (in Japanese) (See Japanese page) 
(in English) Deep learning enables high-accuracy image classification, and different learning tasks may acquire distinct visual capabilities. This study aims to analyze visual capabilities acquired by neural networks trained through reinforcement learning, focusing on active adaptation processes similar to human behavior. We construct 3D environments in Unreal Engine and employ DQN with ResNet18 for visual feature extraction. Agents learn navigation policies by using RGB images and goal position to reach the goal while avoiding obstacles. This study clarifies the process by which action-driven learning progressively acquires visual features.
Keyword (in Japanese) (See Japanese page) 
(in English) Reinforcement Learning / Deep Learning / Visual Representation Learning / Depth Estimation / DQN / / /  
Reference Info. IEICE Tech. Rep., vol. 125, no. 424, PRMU2025-56, pp. 79-83, March 2026.
Paper # PRMU2025-56 
Date of Issue 2026-03-17 (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 PRMU2025-56

Conference Information
Committee PRMU IPSJ-CVIM IBISML ITE-SIP  
Conference Date 2026-03-24 - 2026-03-25 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To PRMU 
Conference Code 2026-03-PRMU-CVIM-IBISML-SIP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Analysis of the Visual Capabilities of a Deep Neural Network Trained via Reinforcement Learning for Autonomous Behavior in a Virtual Environment 
Sub Title (in English)  
Keyword(1) Reinforcement Learning  
Keyword(2) Deep Learning  
Keyword(3) Visual Representation Learning  
Keyword(4) Depth Estimation  
Keyword(5) DQN  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Naoko Kaneda  
1st Author's Affiliation Toyo University graduate school (Toyo univ)
2nd Author's Name Sena Sato  
2nd Author's Affiliation Toyo University (Toyo univ)
3rd Author's Name Makoto murakami  
3rd Author's Affiliation Toyo University (Toyo univ)
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 2026-03-25 09:20:00 
Presentation Time 15 minutes 
Registration for PRMU 
Paper # PRMU2025-56 
Volume (vol) vol.125 
Number (no) no.424 
Page pp.79-83 
#Pages 5 
Date of Issue 2026-03-17 (PRMU) 

[Return to Top Page]

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