IEICE Technical Committee Submission System
Conference Schedule
Online Proceedings
[Sign in]
Tech. Rep. Archives
    [Japanese] / [English] 
( Committee/Place/Topics  ) --Press->
 
( Paper Keywords:  /  Column:Title Auth. Affi. Abst. Keyword ) --Press->

All Technical Committee Conferences  (Searched in: All Years)

Search Results: Conference Papers
 Conference Papers (Available on Advance Programs)  (Sort by: Date Descending)
 Results 1 - 12 of 12  /   
Committee Date Time Place Paper Title / Authors Abstract Paper #
RCC, ISEC, IT, WBS 2024-03-13
- 2024-03-14
Osaka Osaka Univ. (Suita Campus) Construction of an MDL Estimator with Tight Risk Bound in Simple ReLU Networks
Yoshinari Takeishi, Jun'ichi Takeuchi (Kyushu Univ.) IT2023-134 ISEC2023-133 WBS2023-122 RCC2023-116
We consider the problem of estimating the parameters of the final layer in two-layer neural networks with ReLU activatio... [more] IT2023-134 ISEC2023-133 WBS2023-122 RCC2023-116
pp.368-373
PRMU, IBISML, IPSJ-CVIM 2024-03-03
16:30
Hiroshima Hiroshima Univ. Higashi-Hiroshima campus
(Primary: On-site, Secondary: Online)
Assessment of the Utility of Tumor Location Information in MR Image Classification of Tumors
Tsukasa Nishinakagawa, Yoshinari Takeishi, Jun'ichi Takeuchi (Kyushu Univ.) IBISML2023-43
MRI, or magnetic resonance imaging, is a medical imaging technique widely used in various healthcare settings. It utiliz... [more] IBISML2023-43
pp.21-28
IBISML 2023-12-20
15:20
Tokyo National Institute of Informatics
(Primary: On-site, Secondary: Online)
Stabilization and Acceleration of Stochastic Gradient Descent Based on Eigenvalue Decomposition of the Fisher Information Matrix
Masazumi Iida, Yoshinari Takeishi (Kyushu Univ.), Siyang Wang (Umea Univ.), Jun'ichi Takeuchi (Kyushu Univ.) IBISML2023-32
By using approximated eigen-decomposition of the Fisher information matrix, the Stabilizer method was recently developed... [more] IBISML2023-32
pp.13-17
ICSS, IPSJ-SPT 2023-03-13
15:20
Okinawa Okinawaken Seinenkaikan
(Primary: On-site, Secondary: Online)
Multimodal Feature Integration Toward High-Accuracy Network Intrusion Detection
Masazumi Iida, Kohei Miyamoto, Yoshinari Takeishi, Shota Kawanaka (Kyushu Univ.), Chansu Han, Tao Ban, Takeshi Takahashi (NICT), Jun'ichi Takeuchi (Kyushu Univ.) ICSS2022-55
(To be available after the conference date) [more] ICSS2022-55
pp.43-48
MBE, NC 2022-12-03
15:25
Osaka Osaka Electro-Communication University Stabilization and Acceleration of Gradient Descent Based on Eigenvalue Decomposition of the Fisher Information Matrix
Jun'ichi Takeuchi, Yoshinari Takeishi, Masazumi Iida (Kyushu Univ.), Noboru Murata (Waseda Univ.), Kazushi Mimura (Hiroshima City Univ.), Hiroshi Nagaoka (Univ. of Electro Communication) MBE2022-39 NC2022-61
We propose a method to stabilize the gradient decent method without decreasing learning rate for two-layer neural networ... [more] MBE2022-39 NC2022-61
pp.80-85
IBISML 2022-09-15
14:40
Kanagawa Keio Univ. (Yagami Campus)
(Primary: On-site, Secondary: Online)
On Analysis of Generalization Error of Neural Networks
Jun'ichi Takeuchi (Kyushu Univ.), Noboru Murata (Waseda Univ.), Kazushi Mimura (Hiroshima City Univ.), Yoshinari Takeishi, Masazumi Iida (Kyushu Univ.), Hiroshi Nagaoka (Univ. of Electro Communication) IBISML2022-37
 [more] IBISML2022-37
pp.31-38
NC, IBISML, IPSJ-BIO, IPSJ-MPS [detail] 2022-06-27
16:20
Okinawa
(Primary: On-site, Secondary: Online)
Relaxing the Requirement for Decomposition of the Fisher Information Matrix for Simple Softplus Neural Networks
Masazumi Iida, Yoshinari Takeishi, Jun'ichi Takeuchi (Kyushu Univ.) NC2022-6 IBISML2022-6
(To be available after the conference date) [more] NC2022-6 IBISML2022-6
pp.53-60
IT, EMM 2022-05-17
13:00
Gifu Gifu University
(Primary: On-site, Secondary: Online)
On Structure of Fisher Information Matrix for Simple Softplus Neural Networks
Masazumi Iida, Yoshinari Takeishi, Jun'ichi Takeuchi (Kyushu Univ.) IT2022-1 EMM2022-1
 [more] IT2022-1 EMM2022-1
pp.1-6
RCS, SIP, IT 2022-01-21
09:25
Online Online Approximate Eigenvalue Decomposition of Fisher Information Matrix for Simple ReLU Networks
Yoshinari Takeishi, Masazumi Iida, Jun'ichi Takeuchi (Kyushu Univ.) IT2021-68 SIP2021-76 RCS2021-236
We argue the Fisher information matrix (FIM) of one hidden layer networks with the ReLU activation function. Let $W$ den... [more] IT2021-68 SIP2021-76 RCS2021-236
pp.225-230
RCS, IT, SIP 2016-01-18
14:30
Osaka Kwansei Gakuin Univ. Osaka Umeda Campus An Improved Upper Bound on Block Error Probability of Least Squares Sparse Superposition Codes with Bernoulli Dictionary
Yoshinari Takeishi, Jun'ichi Takeuchi (Kyshu Univ.) IT2015-66 SIP2015-80 RCS2015-298
 [more] IT2015-66 SIP2015-80 RCS2015-298
pp.107-112
IT 2012-09-28
10:15
Gunma Kusatsu Seminar House Sparse Superposition Codes with Bernoulli Dictionary
Yoshinari Takeishi, Masanori Kawakita, Jun'ichi Takeuchi (Kyushu Univ.) IT2012-38
We study some extension of the sparse superposition codes proposed by Barron and Joseph in 2010. The original sparse sup... [more] IT2012-38
pp.41-46
PRMU, IBISML, IPSJ-CVIM [detail] 2011-09-05
10:30
Hokkaido   On Evaluation of Stochastic Complexity based on Bayes Code and Its Applications to Model Selection
Yoshinari Takeishi, Masanori Kawakita, Jun'ichi Takeuchi (Kyushu Univ./ISIT) PRMU2011-59 IBISML2011-18
We evaluate stochastic complexity of Gaussian mixture by Bayes code length, and apply it to the model selection problem.... [more] PRMU2011-59 IBISML2011-18
pp.9-14
 Results 1 - 12 of 12  /   
Choose a download format for default settings. [NEW !!]
Text format pLaTeX format CSV format BibTeX format
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)


[Return to Top Page]

[Return to IEICE Web Page]


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