Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML |
2011-11-10 15:45 |
Nara |
Nara Womens Univ. |
Image segmentation and restoration by variational Bayesian method and MCMC Kenta Kayano (Kansai Univ.), Kenji Nagata, Masato Okada (Univ. of Tokyo), Seiji Miyoshi (Kansai Univ.) IBISML2011-68 |
In this paper, we derive a deterministic algorithm that restores and segments an image by using variational Bayesian met... [more] |
IBISML2011-68 pp.175-180 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2011-09-05 10:00 |
Hokkaido |
|
Global Solution of Variational Bayesian Matrix Factorization Under Matrix-wise Independence Shinichi Nakajima (Nikon), Masashi Sugiyama (Tokyo Inst. of Tech.), Derin Babacan (Illinois Univ.) PRMU2011-58 IBISML2011-17 |
Variational Bayesian matrix factorization (VBMF) efficiently
approximates the posterior distribution of factorized mat... [more] |
PRMU2011-58 IBISML2011-17 pp.1-8 |
NC |
2011-07-26 11:00 |
Hyogo |
Graduate School of Engineering, Kobe University |
Image Segmentation and Restoration using Region-Based Hidden Variables and Belief Propagation Ryota Hasegawa (Kansai Univ.), Masato Okada (Univ. of Tokyo), Seiji Miyoshi (Kansai Univ.) NC2011-35 |
We derive a deterministic algorithm that restores and segments an image using belief propagation and a variational Bayes... [more] |
NC2011-35 pp.81-86 |
IBISML |
2011-03-29 10:30 |
Osaka |
Nakanoshima Center, Osaka Univ. |
On Automatic Dimensionality Selection in Probabilistic PCA Shinichi Nakajima (Nikon), Masashi Sugiyama (Tokyo Inst. of Tech./JST), Derin Babacan (Illinois of.Univ.) IBISML2010-123 |
In probabilistic PCA,
the fully Bayesian estimation is computationally intractable.
To cope with this problem,
two ty... [more] |
IBISML2010-123 pp.131-138 |
NC, MBE (Joint) |
2011-03-08 10:40 |
Tokyo |
Tamagawa University |
Effect of Information Source on Cross Validation in Variational Bayes Learning Shinji Oyama, Sumio Watanabe (Tokyo Tech.) NC2010-167 |
The variational Bayes learning provides high generalization performance as the Bayes learning using a small computationa... [more] |
NC2010-167 pp.235-240 |
IBISML |
2010-11-05 15:30 |
Tokyo |
IIS, Univ. of Tokyo |
[Poster Presentation]
On the difference between Bayes and Variational Bayes in a normal mixture Tetsutaro Yamada, Sumio Watanabe (Tokyo Tech.) IBISML2010-84 |
Variational Bayes method or mean field approximation
is widely used because it enables us to estimate the information s... [more] |
IBISML2010-84 pp.181-186 |
IBISML, PRMU, IPSJ-CVIM [detail] |
2010-09-06 09:00 |
Fukuoka |
Fukuoka Univ. |
A Study of Variances of Cross-Validation and Generalization Error in Variational Bayes Method Shinji Oyama, Sumio Watanabe (Tokyo Tech.) PRMU2010-74 IBISML2010-46 |
Variational Bayes method provides high generalization performance as the Bayes method using a small computational cost a... [more] |
PRMU2010-74 IBISML2010-46 pp.135-142 |
NC |
2010-07-27 14:55 |
Kyoto |
Kyoto University |
Hierarchical Bayes method for NIRS-DOT inverse problem and its phase diagrams Atsushi Miyamoto, Kazuho Watanabe, Kazushi Ikeda (NAIST), Masa-aki Sato (ATR) NC2010-38 |
The NIRS-DOT is a method to reconstruct tomographic images from the data by solving the linear equations, which have amb... [more] |
NC2010-38 pp.51-56 |
NC, MBE (Joint) |
2010-03-10 14:10 |
Tokyo |
Tamagawa University |
Variational Bayes approach for NIRS-DOT inverse problem Atsushi Miyamoto, Kazuho Watanabe, Kazushi Ikeda (NAIST), Masa-aki Sato (ATR) NC2009-137 |
NIRS-DOT is tomography which is reconstructed from NIRS data. NIRS-DOT requires to solve the inverse problem which has i... [more] |
NC2009-137 pp.291-296 |
PRMU, SP, MVE, CQ |
2010-01-21 11:40 |
Kyoto |
Kyoto Univ. |
Online speaker clustering using an ergodic HMM and its application to meeting minute generation Takafumi Koshinaka, Kentaro Nagatomo, Kenji Satoh (NEC Corp.) CQ2009-62 PRMU2009-161 SP2009-102 MVE2009-84 |
A novel online speaker clustering method suitable for real-time applications is proposed. Using an ergodic hidden Marko... [more] |
CQ2009-62 PRMU2009-161 SP2009-102 MVE2009-84 pp.39-44 |
NC, MBE (Joint) |
2009-03-13 09:20 |
Tokyo |
Tamagawa Univ. |
Modular Reinforcement Learning based on Adaptive Model Complexity Yu Hiei (Nara Inst. of Sci and Tech.), Takeshi Mori (Kyoto Univ.), Shin Ishii (Kyoto Univ./Nara Institute of Science and Technology) NC2008-149 |
In real-world problems such as robot control, the environment surrounding a controlled system is nonstationary, and the ... [more] |
NC2008-149 pp.273-278 |
PRMU |
2009-02-20 11:15 |
Tokyo |
Univ. of Tokyo (IIS) |
Automatically Estimating Number of Scenes for Video Summarization using Model Selection Criteria Koji Yamasaki, Koichi Shinoda, Sadaoki Furui (Tokyo Inst. of Tech.) PRMU2008-231 |
This paper describes a video summarization system using model selection techniques to estimate the optimal number of sce... [more] |
PRMU2008-231 pp.139-144 |
NC |
2009-01-19 13:55 |
Hokkaido |
Hokkaido Univ. |
On the Effect of Hyperparameter to Generalization Error in Variational Bayes Learning Shinji Oyama, Sumio Watanabe (Tokyo Inst. of Tech.) NC2008-87 |
In variational Bayes learning, the probability distribution of the hidden variable and parameter is made by the mean fie... [more] |
NC2008-87 pp.31-36 |
NC, MBE (Joint) |
2008-12-20 10:00 |
Aichi |
Nagoya Inst. Tech. |
Clustering complex networks with the prior based on degree distribution Naoyuki Harada, Ichiro Takeuchi (NIT), Ryohei Nakano (Chubu Univ.) NC2008-73 |
Newman et al. proposed a graph clustering method based on a robabilistic mixture model with only the general assumption ... [more] |
NC2008-73 pp.1-6 |
NC, MBE (Joint) |
2008-12-20 11:05 |
Aichi |
Nagoya Inst. Tech. |
A Study on Variational Bayes Method with the Primitive Initial Point Yuta Ishikawa, Ichiro Takeuchi (NIT), Ryohei Nakano (Chubu Univ.) NC2008-75 |
The variational bayes (VB) method is widely used as an approximation of
the bayes method.
Since the objective functio... [more] |
NC2008-75 pp.13-18 |
HIP |
2008-03-22 - 2008-03-23 |
Fukuoka |
Kitakyushu Science and Research Park |
Object Concept Modeling Based on Relationship among Visual Information, Usage and Function Yasuhito Shinchi, Takayuki Nagai (UEC) HIP2007-159 |
A novel object concept model, which encodes relationship among visual information, function and usage, is proposed in th... [more] |
HIP2007-159 pp.1-6 |
SIS |
2008-03-14 13:10 |
Tokyo |
Musashi Institute of Technology(Setagaya) |
A Model of Tools Usage Based on Visual Observation Yasuhito Shichi, Takayuki Nagai (UEC) SIS2007-86 |
Recognizing object function is a very important aspect of object recognition tasks as well as recognizing object visuall... [more] |
SIS2007-86 pp.27-32 |
NC |
2007-10-18 09:55 |
Miyagi |
Tohoku University |
Variational Bayes Hidden Markov Models for extracting spatiotemporal spike pattern Kentaro Katahira (Univ. Tokyo/RIKEN), Jun Nishikawa, Kazuo Okanoya (RIKEN), Masato Okada (Univ. Tokyo/RIKEN) NC2007-34 |
Hidden Markov Model (HMM) is used to extracting spatio-temporal pattern from spikes recorded by
multielectrode. The EM ... [more] |
NC2007-34 pp.7-12 |
NC |
2007-03-14 15:10 |
Tokyo |
Tamagawa University |
Markov and semi-Markov switching of source appearances for non-stationary independent component analysis Junichiro Hirayama (NAIST/JSPS), Shin-ichi Maeda, Shin Ishii (NAIST) |
Independent Component Analysis (ICA) is currently the most popularly used approach to blind source separation (BSS), the... [more] |
NC2006-147 pp.173-178 |
NC |
2007-03-15 15:30 |
Tokyo |
Tamagawa University |
Deterministic Annealing in Variational Baysian Algorithm Kentaro Katahira (Univ. Tokyo/RIKEN), Kazuho Watanabe (Tokyo Tech), Masato Okada (Univ. Tokyo/RIKEN) |
Variational Bayes (VB) algorithm is widely used as an approximation of Bayesian method. The VB algorithm can approximate... [more] |
NC2006-183 pp.177-182 |