Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Model Selection of Layered Neural Networks using WBIC based on Steepest Descent and MCMC Method Yusuke Tamai, Sumio Watanabe (Tokyo Inst. of Tech.) IBISML2013-36 |
Many learning machines such as neural networks, normal mixtures, and hidden Markov Models contain hierarchical layers, h... [more] |
IBISML2013-36 pp.1-6 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Global Solvers for Variational Bayesian Low-rank Subspace Clustering Shinichi Nakajima (Nikon), Akiko Takeda (Univ. of Tokyo), S. Derin Babacan (Google), Masashi Sugiyama (Tokyo Inst. of Tech.), Ichiro Takeuchi (Nagoya Inst. of Tech.) IBISML2013-37 |
Variational Bayesian (VB) learning, known to be a promising approximation method to Bayesian learning,
is generally per... [more] |
IBISML2013-37 pp.7-14 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
A boosting method considering tolerance against noisy data by weighting each data according to the distance between incidents Shinjiro Fujita, Sayaka Kamei, Satoshi Fujita (Hiroshima Univ.) IBISML2013-38 |
AdaBoost is one of the major ensemble learning methods. It is easy to implement and
has high classification accuracy. ... [more] |
IBISML2013-38 pp.15-21 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Safe Sample Screening Rule on Hinge Loss Minimization Kohei Ogawa, Yamato Kawamoto, Yoshiki Suzuki, Ichiro Takeuchi (Nagoya Inst. of Tech.) IBISML2013-39 |
In this paper, we propose the algorithm that can speed up computing problems minimizing Hinge loss such as SVMs, via eli... [more] |
IBISML2013-39 pp.23-30 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Statistical mechanics of Bayesian optimal dictionary learning Ayaka Sakata, Yoshiyuki Kabashima (Tokyo Inst. of Tech.) IBISML2013-40 |
Dictionary learning is a problem to learn a dictionary matrix D (M times N dimension) and a sparse matrix X (N times P ... [more] |
IBISML2013-40 pp.31-38 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Performance Comparisons between Dependency Networks and Bayesian Networks Kazuya Takabatake, Shotaro Akaho (AIST) IBISML2013-41 |
Dependency networks are graphical models in which tasks of learning are done by totally local and simple algorithms of i... [more] |
IBISML2013-41 pp.39-44 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Optimization-based Multivariate Two-sample Test and Its Efficient Computation Yuki Shinmura, Yusuke Saida, Ichiro Takeuchi (Nagoya Inst. of Tech.) IBISML2013-42 |
Multivariate two-sample test has recently attracted a great deal of interest and its importance has been increasing in m... [more] |
IBISML2013-42 pp.45-52 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Direct Conditional Probability Density Estimation based on Sparse Additive Models Motoki Shiga (Gifu Univ.), Masashi Sugiyama (Tokyo Inst. of Tech.) IBISML2013-43 |
On identification of the statistical dependency between inputs and outputs, an conditional density estimation is essenti... [more] |
IBISML2013-43 pp.53-60 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Exaluation of Revised IP-OLDF with S-SVM, LDF and logistic regression by K-fold cross-validation Shuichi Shinmura (Seikei Univ.) IBISML2013-44 |
In this paper, Revised IP-OLDF based on MNM criterion is proposed using a mixed integer programming. The new discriminan... [more] |
IBISML2013-44 pp.61-68 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Non-Convex Optimization of Robust Support Vector Regression by Utilizing Parametric Programming Shinya Suzumura, Ichiro Takeuchi (Nagoya Inst. of Tech.), Masashi Sugiyama (Tokyo Inst. of Tech.) IBISML2013-45 |
In this paper, we propose a novel optimization method for robust support vector regression(SVR) that has robustness to o... [more] |
IBISML2013-45 pp.69-75 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Query auditing for privacy preserving similarity search Hiromi Arai (RIKEN), Koji Tsuda (AIST), Jun Sakuma (Univ. of Tsukuba) IBISML2013-46 |
In this paper, we propose a query auditing method for similarity searches that examines whether database responces satis... [more] |
IBISML2013-46 pp.77-83 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Inter-subject Common Spatial Pattern Filter for Motor Imagery based Brain Computer Interface Tatsuya Yokota, Yukihiko Yamashita (Tokyo Inst. of Tech.), Andrzej Cichocki (RIKEN) IBISML2013-47 |
Motor imagery based brain computer interface is a technique to control some device (e.g., wheelchair) through motor imag... [more] |
IBISML2013-47 pp.85-90 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Typical Performance of Sharpe Ratio Using Replica Analysis Takashi Shinzato (Akita Prefectural Univ.) IBISML2013-48 |
[more] |
IBISML2013-48 pp.91-98 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Predicting Protein Complexes by Sampling More Accurately and Efficiently Chasanah Kusumastuti Widita, Osamu Maruyama (Kyushu Univ.) IBISML2013-49 |
The problem of predicting sets of components of heteromeric protein
complexes is a challenging problem in Systems Bio... [more] |
IBISML2013-49 pp.99-106 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
A Study On Multiple Matrix Factorization Masahiro Kohjima, Kenji Esaki, Noriko Takaya, Hiroshi Sawada (NTT) IBISML2013-50 |
In this study, we propose new matrix factorization methods for multiple matrices. The research to analyze multiple data ... [more] |
IBISML2013-50 pp.107-114 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Support vector comparison machines Toby Dylan Hocking, Supaporn Spanurattana, Masashi Sugiyama (Tokyo Inst. of Tech.) IBISML2013-51 |
In ranking problems, the goal is to learn a ranking function
$r(x)inRR$ from labeled pairs $x,x'$ of input points. In... [more] |
IBISML2013-51 pp.115-121 |
IBISML |
2013-11-12 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
The Method to Extract Latent Skills from Time Series Examination Results with Matrix Factorization Shinichi Oeda, Eriko Amano (KNCT), Kenji Yamanishi (Univ. of Tokyo) IBISML2013-52 |
Examination results are used to judge whether an student possesses desired latent skills. In order to grasp the skills, ... [more] |
IBISML2013-52 pp.123-130 |
IBISML |
2013-11-13 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Computationally Efficient Estimation of Squared-loss Mutual Information with Multiplicative Kernel Models Tomoya Sakai, Masashi Sugiyama (Tokyo Inst. of Tech.) IBISML2013-53 |
emph{Squared-loss mutual information} (SMI) is a robust measure of statistical dependence between random variables.
The... [more] |
IBISML2013-53 pp.131-137 |
IBISML |
2013-11-13 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Sample Complexity Reduction in Reinforcement Learning by Transferred Transition and Reward Probability Kouta Oguni, Kazuyuki Narisawa, Ayumi Shinohara (Tohoku Univ.) IBISML2013-54 |
Most existing reinforcement learning algorithms are not very efficient in real environmental problems. Because, they hav... [more] |
IBISML2013-54 pp.139-146 |
IBISML |
2013-11-13 15:45 |
Tokyo |
Tokyo Institute of Technology, Kuramae-Kaikan |
[Poster Presentation]
Distributional Statistics Estimation via Kernel Mean Embeddings
-- Density Function, Credible interval, and Moment Estimation -- Motonobu Kanagawa (SOKENDAI), Kenji Fukumizu (ISM) IBISML2013-55 |
The RKHS embedding approach for nonparametric statistical inference, in which probability distributions are represented ... [more] |
IBISML2013-55 pp.147-154 |