Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
SR |
2024-05-21 10:30 |
Kagoshima |
Yokacenter (Kagoshima) (Primary: On-site, Secondary: Online) |
[Short Paper]
Distributed Learning with Deep Joint Source Channel Coding for Overfitting Avoidance Issa Matsumura, Katsuya Suto (UEC) |
(To be available after the conference date) [more] |
|
MI |
2024-03-04 09:36 |
Okinawa |
OKINAWAKEN SEINENKAIKAN (Primary: On-site, Secondary: Online) |
[Short Paper]
Overfitting Prevention for PET Image Reconstruction using Early Stopping of Deep Image Prior based on Unbiased Risk Estimator Kaito Matsumura, Hidekata Hontani (NIT), Muneyuki Sakata (TMIG), Yuichi Kimura (KDU), Tatsuya Yokota (NIT) MI2023-65 |
In recent years, methods for PET image reconstruction using Deep Image Prior (DIP) have been actively studied. In PET im... [more] |
MI2023-65 pp.106-108 |
IE, ITS, ITE-MMS, ITE-ME, ITE-AIT [detail] |
2023-02-22 09:15 |
Hokkaido |
Hokkaido Univ. |
Varying Difficulties in the Data and Its Effects to the Generalizability Aoshi Kawaguchi (UTokyo), Hiroshi Kera (Chiba University), Toshihiko Yamasaki (UTokyo) ITS2022-57 IE2022-74 |
Deep neural networks (DNNs) often undergo overfitting.
One cause of overfitting is the training time.
It was proven th... [more] |
ITS2022-57 IE2022-74 pp.83-88 |
PN |
2022-08-29 09:55 |
Hokkaido |
(Primary: On-site, Secondary: Online) |
Comparison of overfitting in ANN- and VSTF-based nonlinear equalizers to repeated random bit patterns Kai Ikuta, Nakamura Jinya, Motai Daisuke, Nakamura Moriya (Meiji Univ.) PN2022-9 |
We compared the overfitting characteristics of artificial-neural-network- (ANN-) and Volterra-series-transfer-function- ... [more] |
PN2022-9 pp.5-9 |
NC, IBISML, IPSJ-BIO, IPSJ-MPS [detail] |
2022-06-27 14:25 |
Okinawa |
(Primary: On-site, Secondary: Online) |
A Bagging Method to Improve the Accuracy of Gaussian Process Regression for Neural Architecture Search Rion Hada, Masao Okita, Fumihiko Ino (Osaka Univ.) NC2022-2 IBISML2022-2 |
The goal of this study is to improve performance estimation for neural network architectures in neural architecture sear... [more] |
NC2022-2 IBISML2022-2 pp.6-13 |
SP |
2011-07-21 15:00 |
Hokkaido |
Jozankei Grand Hotel |
Construction of Speaker Model Using A New GMM Learning Method Based on Clustering Masaki Mifune, Motoyuki Suzuki, Fuji Ren, Kenji Kita (Univ. of Tokushima) SP2011-42 |
In the speaker identification research fields,
Gaussian Mixture Models (GMM) are widely used as speaker models because ... [more] |
SP2011-42 pp.7-10 |
NC |
2007-05-21 10:25 |
Kanagawa |
Tokyo Inst. Tech.(Suzukakedai Campus) |
Unbiased Likelihood Backpropagation Learning Masashi Sekino, Katsumi Nitta (Tokyo Inst. of Tech.) NC2007-1 |
The error backpropagation is one of the popular methods for training an artificial neural network.When the error backpro... [more] |
NC2007-1 pp.1-6 |
NC |
2007-03-14 10:10 |
Tokyo |
Tamagawa University |
Unbiased Learning for Hierarchical Models Masashi Sekino, Katsumi Nitta (Tokyo Tech) |
It is known that overfitting occurs when a conventional statistical learning method such as maximum likelihood estimatio... [more] |
NC2006-136 pp.109-114 |