Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
NC, IBISML, IPSJ-BIO, IPSJ-MPS [detail] |
2023-06-30 11:10 |
Okinawa |
OIST Conference Center (Primary: On-site, Secondary: Online) |
On performance degradation of a method by minimizing the conditional mutual information for the out-of-distribution generalization Genki Takahashi, Toshiyuki Tanaka (Kyoto University) NC2023-15 IBISML2023-15 |
In the out-of-distribution generalization problem, the smaller the degree of change in the data generating distribution ... [more] |
NC2023-15 IBISML2023-15 pp.91-97 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2023-03-03 16:55 |
Hokkaido |
Future University Hakodate (Primary: On-site, Secondary: Online) |
Upper bound of real log canonical threshold based on linear programming problem for the multi-indexes of a polynomial Joe Hirose (Tokyo Tech) PRMU2022-125 IBISML2022-132 |
A real log canonical threshold (RLCT) is an invariant which gives a Bayesian generalization error. While a strict value ... [more] |
PRMU2022-125 IBISML2022-132 pp.363-370 |
IBISML |
2022-12-22 15:10 |
Kyoto |
Kyoto University (Primary: On-site, Secondary: Online) |
Effect of Prior Distribution when Data-Generating Process is in a Neighborhood of Singularities of Learning Machines Nozomi Maki, Sumio Watanabe (TokyoTech) IBISML2022-46 |
Learning machines which have hierarchical structure or latent variables such as deep learning or normal mixtures contain... [more] |
IBISML2022-46 pp.18-23 |
IBISML |
2022-03-08 10:55 |
Online |
Online |
Real log canonical threshold of reduced rank regression when inputs are on a low dimensional hyperplane Joe Hirose, Sumio Watanabe (Tokyo Tech) IBISML2021-32 |
A reduced rank regression is a statistical model which estimates a linear regression function from in- puts to outputs w... [more] |
IBISML2021-32 pp.15-18 |
RCS |
2021-06-24 13:00 |
Online |
Online |
A Study on Recurrent Neural Network Aided GNSS Positioning Kohei Nishioka, Shinsuke Ibi (Doshisha Univ.), Takumi Takahashi (Osaka Univ.), Hisato Iwai (Doshisha Univ.) RCS2021-53 |
One method of positioning schemes with the aid of the global navigation satellite system (GNSS) is to approximately solv... [more] |
RCS2021-53 pp.145-150 |
SIS |
2021-03-04 15:50 |
Online |
Online |
[Invited Talk]
A Proposal in Order to use AI Safely for Control Nobuyasu Kanekawa (Hitachi) SIS2020-50 |
The research of the deep learning which developed the "neural Network" proposed by the second AI (artificial intelligenc... [more] |
SIS2020-50 pp.83-87 |
MBE, NC, NLP, CAS (Joint) [detail] |
2020-10-30 10:50 |
Online |
Online |
statistical mechanical analysis of catastrophic forgetting in continual learning with teacher and student networks Haruka Asanuma, Shiro Takagi, Yoshihiro Nagano, Yuki Yoshida (Tokyo Univ.), Yasuhiko Igarashi (Tsukuba Univ.), Masato Okada (Tokyo Univ.) NC2020-18 |
When single neural networks sequentially learns more than one task, catastrophic forgetting occurs except for the last t... [more] |
NC2020-18 pp.50-55 |
IBISML |
2020-01-09 13:00 |
Tokyo |
ISM |
Asymptotic Behavior of Bayesian Generalization Error in Multinomial Mixtures Takumi Watanabe, Sumio Watanabe (Tokyo Tech) IBISML2019-18 |
Multinomial mixtures are widely used in the information engineering field. However, it is not subject to the conventiona... [more] |
IBISML2019-18 pp.1-8 |
IBISML |
2020-01-09 13:25 |
Tokyo |
ISM |
Real Log Canonical Threshold of Three Layered Neural Network with Swish Activation Function Raiki Tanaka, Sumio Watanabe (Tokyo Tech) IBISML2019-19 |
In neural network learning, it is known that selection of activation function effects generalization performance. Althou... [more] |
IBISML2019-19 pp.9-15 |
IBISML |
2016-11-16 15:00 |
Kyoto |
Kyoto Univ. |
[Poster Presentation]
Optimization Method of Deep Ensemble Learning using Hierarchical Clustering Natsuki Koda, Sumio Watanabe (Tokyo Tech) IBISML2016-70 |
The method which is used for prediction by combining many different learning machines generated by using same training d... [more] |
IBISML2016-70 pp.171-176 |
NC, NLP (Joint) |
2016-01-29 15:50 |
Fukuoka |
Kyushu Institute of Technology |
Node-perturbation Learning for Soft-committee machine Kazuyuki Hara (Nihon Univ.), Kentaro Katahira (Nagoya Univ.) NC2015-66 |
Node perturbation learning is a stochastic gradient descent method for neural networks. It estimates the gradient of the... [more] |
NC2015-66 pp.49-54 |
NLP |
2015-07-22 11:00 |
Hokkaido |
Bibai Onsen Yu-rinkan |
Optimization of Generalization Based on Genetic Algorithm
-- Application for Function Analysis and Pattern Recognition -- Hideki Satoh (Future Univ. Hakodate), Daisuke Kasai (Tokachi Foundation), Masako Satoh (Insent) NLP2015-78 |
A function approximation method was developed taking generalization into account, and it was applied to a pattern recogn... [more] |
NLP2015-78 pp.57-62 |
MICT, WBS |
2014-07-29 12:30 |
Osaka |
Osaka City Univ. (Umeda Satellite) |
[Poster Presentation]
A Study of LED-to-LED Visible Light Communication with Generalization Modified Prime Sequence Code Naoya Murata, Yusuke Kozawa, Yohtaro Umeda (Tokyo Univ. of Science), Hiromasa Habuchi (Ibaraki Univ.) WBS2014-21 MICT2014-35 |
In this paper, an LED-to-LED VLCs (Visible Light Communication) system with generalized modified prime sequence codes (G... [more] |
WBS2014-21 MICT2014-35 pp.67-72 |
NC, MBE (Joint) |
2013-07-19 14:30 |
Tokushima |
The University of Tokushima |
Statistical Mechanics of node-perturbation Learning using two independent noises Kazuyuki Hara (Nihon Univ.), Kentaro Katahira, Masato Okada (Univ. of Tokyo) NC2013-17 |
Node perturbation learning is a stochastic gradient descent method for neural networks. It estimates the gradient by com... [more] |
NC2013-17 pp.13-18 |
TL |
2012-12-08 16:25 |
Tokyo |
WASEDA University |
The effect of preemption on the transitivity alternations in L2 English Takehiko Yagi, Takaaki Suzuki (Kyoto Sangyo Univ.) TL2012-46 |
This study investigates the effect of preemption on the transitivity alternations in second language acquisition of Engl... [more] |
TL2012-46 pp.71-76 |
IBISML |
2011-11-10 15:45 |
Nara |
Nara Womens Univ. |
Nonparametric Estimation of Mixture Model and Minimum Divergence Methods Kazuho Watanabe (NAIST), Shiro Ikeda (ISM) IBISML2011-78 |
We discuss a nonparametric estimation method of the mixing distribution in mixture models.
We propose an objective fun... [more] |
IBISML2011-78 pp.243-249 |
DC |
2011-10-20 13:30 |
Tokyo |
|
Neighborhood Level Error Control Codes for Multiple-level Systems Shohei Kotaki, Masato Kitakami (Chiba Univ.) DC2011-23 |
Multiple-level concept, which deals with more than 2 states as data unit, is used in such systems as Flash Memory or mod... [more] |
DC2011-23 pp.19-24 |
NC |
2011-10-20 13:10 |
Fukuoka |
Ohashi Campus, Kyushu Univ. |
Statistical Mechanics of Node-Perturbation Learning for Nonlinear Perceptron Kazuyuki Hara (Nihon Univ.), Kentaro Katahira (JST), Kazuo Okanoya (RIKEN), Masato Okada (Tokyo Univ.) NC2011-63 |
Node-perturbation learning is a kind of statistical gradient descent algorithm that can be applied to problems where the... [more] |
NC2011-63 pp.107-112 |
NC, IPSJ-BIO [detail] |
2011-06-24 14:15 |
Okinawa |
50th Anniversary Memorial Hall, University of the Ryukyus |
The Capability of Selective Desensitization Neural Networks at Two-Variable Function Approximation Kazuaki Nonaka, Fumihide Tanaka, Masahiko Morita (Tsukuba Univ) NC2011-14 |
Selective Desensitization Neural Network (SDNN) is known to be able to approximate some functions well with high general... [more] |
NC2011-14 pp.113-118 |
IBISML |
2011-03-29 16:30 |
Osaka |
Nakanoshima Center, Osaka Univ. |
Fast Convergence Rate of Multiple Kernel Learning with Elastic-net Regularization Taiji Suzuki, Ryota Tomioka (Univ. of Tokyo), Masashi Sugiyama (Tokyo Inst. of Tech.) IBISML2010-126 |
We investigate the learning rate of multiple kernel leaning (MKL)
with elastic-net regularization,
which consists of a... [more] |
IBISML2010-126 pp.153-160 |