IEICE Technical Report

Print edition: ISSN 0913-5685      Online edition: ISSN 2432-6380

Volume 114, Number 306

Infomation-Based Induction Sciences and Machine Learning

[PREV] [NEXT]

[TOP] | [2011] | [2012] | [2013] | [2014] | [2015] | [2016] | [2017] | [Japanese] / [English]

[PROGRAM] [BULK PDF DOWNLOAD]

Table of contents

IBISML2014-35
[Poster Presentation] Application of partial parallel interference cancellation to sparse signal recovery
Koujin Takeda (Ibaraki Univ.), Toshiyuki Tanaka (Kyoto Univ.)
pp. 1 - 7

IBISML2014-36
[Poster Presentation] Feature Extraction for Image Classification using Restricted Boltzmann Machines
Reiki Suda, Koujin Takeda (Ibaraki Univ.)
pp. 9 - 15

IBISML2014-37
[Poster Presentation] Approximate Models of Probability Distributions for Principal Components of Sample Mahalanobis Distances -- About Each Element and Partial Sum of Sample Mahalanobis Distances --
Yasuyuki Kobayashi (Teikyo Univ.)
pp. 17 - 24

IBISML2014-38
[Poster Presentation] Theoretical Analysis of Empirical MAP and Empirical Partially Bayes
Shinichi Nakajima (TU Berlin), Masashi Sugiyama (Univ. of Tokyo)
pp. 25 - 32

IBISML2014-39
[Poster Presentation] Entropy Estimators Based on Simple Linear Regression
Hideitsu Hino (Univ. of Tukuba), Kensuke Koshijima (Recruit), Noboru Murata (Waseda Univ.)
pp. 33 - 40

IBISML2014-40
[Poster Presentation] Importance-Weighted Covariance Estimation for Robust Common Spatial Pattern
Alessandro Balzi (PoliMi), Florian Yger, Masashi Sugiyama (Univ. of Tokyo)
pp. 41 - 48

IBISML2014-41
[Poster Presentation] Breakdown Point of Robust Support Vector Machine
Takafumi Kanamori (Nagoya Univ.), Shuhei Fujiwara, Akiko Takeda (Univ. of Tokyo)
pp. 49 - 56

IBISML2014-42
[Poster Presentation] Possibilities and limitations of machine learning on unweighted graphs -- From the viewpoint of random geometric graph theory --
Yoshikazu Terada (NICT), Ulrike von Luxburg (Univ. of Hamburg)
pp. 57 - 64

IBISML2014-43
[Poster Presentation] Exact SVM Training by Wolfe's Minimum Norm Point Algorithm
Masashi Kitamura, Akiko Takeda, Satoru Iwata (Univ. of Tokyo)
pp. 65 - 71

IBISML2014-44
[Poster Presentation] Efficient leave-one-out cross-validation for L2-regularized classifier
Shota Okumura, Yoshiki Suzuki, Kohei Ogawa, Yuki Shinmura, Ichiro Takeuchi (NIT)
pp. 73 - 80

IBISML2014-45
[Poster Presentation] Geometry of Fisher's linear discriminant analysis
Jun Fujiki, Masaru Tanaka (Fukuoka Univ.), Hitoshi Sakano (NTT data), Akisato Kimura (NTT)
pp. 81 - 86

IBISML2014-46
[Poster Presentation] Heuristic principal component analysis based unsupervised feature extraction and its application to bioinformatics
Y-h. Taguchi (Chuo Univ)
pp. 87 - 94

IBISML2014-47
[Poster Presentation] Differential Privacy on Linear Regression Model of Crowdsensing
Tran Quang Khai, Kazuto Fukuchi, Jun Sakuma (Univ. of Tsukuba)
pp. 95 - 102

IBISML2014-48
[Poster Presentation] Reconstructive Neural Net based on Context Free Grammar
Takuo Hamaguchi, Masashi Shimbo, Yuji Matsumoto (NAIST)
pp. 103 - 110

IBISML2014-49
[Poster Presentation] Multitask learning meets tensor factorization: task imputation via convex optimization
Kishan Wimalawarne (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo), Ryota Tomioka (TTIC)
pp. 111 - 118

IBISML2014-50
[Poster Presentation] Statistical mechanical analysis of lossy compression by overcomplete basis
Yoshinori Nakanishi-Ohno (Univ. Tokyo), Tomoyuki Obuchi (Tokyo Tech), Masato Okada (Univ. Tokyo), Yoshiyuki Kabashima (Tokyo Tech)
pp. 119 - 126

IBISML2014-51
[Poster Presentation] Direct Density Ratio Estimation with Convolutional Neural Networks with Application in Outlier Detection
Hyunha Nam (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo)
pp. 127 - 132

IBISML2014-52
[Poster Presentation] Direct Density-Derivative Estimation and Its Application in KL-Divergence Approximation
Hiroaki Sasaki (Univ. of Tokyo), Yung-Kyun Noh (KAIST), Masashi Sugiyama (Univ. of Tokyo)
pp. 133 - 140

IBISML2014-53
[Poster Presentation] An Online Policy Gradient Algorithm for Continuous State and Action Markov Decision Processes with Bandit Feedback
Yao Ma (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo)
pp. 141 - 148

IBISML2014-54
[Poster Presentation] Efficient Method for Computing Belief Propagation, with Application to CDMA Multiuser Detection
Arise Kuriya, Toshiyuki Tanaka (Kyoto Univ)
pp. 149 - 153

IBISML2014-55
[Poster Presentation] Target Shift Adaptation for Conditional Density Estimation and Regression
Tuan Duong Nguyen (Tokyo Inst. of Tech.), Marthinus Christoffel du Plessis, Masashi Sugiyama (Univ. of Tokyo)
pp. 155 - 160

IBISML2014-56
[Poster Presentation] Training Algorithm for Restricted Boltzmann Machines Using Auxiliary Function Approach
Norihiro Takamune (Univ. of Tokyo), Hirokazu Kameoka (Univ. of Tokyo/NTT)
pp. 161 - 168

IBISML2014-57
[Poster Presentation] Unified approach for auditory scene analysis based on multichannel factorial hidden Markov model
Takuya Higuchi (Univ. of Tokyo), Hirokazu Kameoka (Univ. of Tokyo/NTT)
pp. 169 - 176

IBISML2014-58
[Poster Presentation] Regularized multi-task learning for multi-dimensional log-density gradient estimation
Ikko Yamane (Tokyo Inst. of Tech.), Hiroaki Sasaki, Masashi Sugiyama (Univ. of Tokyo)
pp. 177 - 183

IBISML2014-59
[Poster Presentation] Online Direct Density-ratio Estimation under the Kullback-Leibler Loss
Marthinus Christoffel du Plessis (Univ. of Tokyo), Hiroaki Shiino (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo)
pp. 185 - 190

IBISML2014-60
[Poster Presentation] An Infinite Superimposed Discrete All-pole Model based on Gamma Processes for Music Signal Analysis
Kazuyoshi Yoshii, Katsutoshi Itoyama (Kyoto Univ.), Masataka Goto (AIST)
pp. 191 - 198

IBISML2014-61
[Poster Presentation] A note on least angle regression in orthogonal case
Katsuyuki Hagiwara (Mie Univ.)
pp. 199 - 206

IBISML2014-62
[Poster Presentation] Denoising High-dimensional Sequences with the Bidirectional Recurrent Restricted Boltzmann Machine
Shoken Kaneko (Yamaha Co.)
pp. 207 - 212

IBISML2014-63
[Poster Presentation] Optimization Method of LASSO Hyperparameter using WAIC
Dai Miyazaki, Sumio Watanabe (Tokyo Tech)
pp. 213 - 218

IBISML2014-64
[Poster Presentation] Asymptotic Analysis of Variational Bayesian Latent Dirichlet Allocation
Shinichi Nakajima (TU Berlin), Issei Sato, Masashi Sugiyama (Univ. of Tokyo), Kazuho Watanabe (Toyohashi Univ. of Tech.), Hiroko Kobayashi (Nikon)
pp. 219 - 226

IBISML2014-65
[Poster Presentation] Learning from Positive and Unlabeled Data 1: Classifier Training and Theoretical Analysis
Marthinus Christoffel du Plessis (Univ. of Tokyo), Gang Niu (Baidu), Masashi Sugiyama (Univ. of Tokyo)
pp. 227 - 233

IBISML2014-66
[Poster Presentation] Learning from Positive and Unlabeled Data 2: Computationally Efficient Estimation of Class Priors
Marthinus Christoffel du Plessis (Univ. of Tokyo), Gang Niu (Baidu), Masashi Sugiyama (Univ. of Tokyo)
pp. 235 - 241

IBISML2014-67
[Poster Presentation] Parallel Distributed Block Coordinate Descent Methods based on Pairwise Comparison Oracle
Kota Matsui (Nagoya Inst. of Tech.), Wataru Kumagai (Kanagawa Univ.), Takafumi Kanamori (Nagoya Univ.)
pp. 243 - 250

IBISML2014-68
[Poster Presentation] Robust Estimation under Heavy Contamination using Unnormalized Models
Takafumi Kanamori (Nagoya Univ.), Hironori Fujisawa (ISM)
pp. 251 - 258

IBISML2014-69
[Poster Presentation] Unsupervised Dimension Reduction via Least-Squares Quadratic Mutual Information
Janya Sainui (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo)
pp. 259 - 262

IBISML2014-70
[Poster Presentation] Support consistency of direct sparse-change learning in Markov networks
Song Liu, Taiji Suzuki (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo)
pp. 263 - 269

IBISML2014-71
[Poster Presentation] Coping with Class Balance Change in Classification : Class-Prior Estimation with Energy Distance
Hideko Kawakubo (Tokyo Inst. of Tech.), Marthinus Christoffel du Plessis, Masashi Sugiyama (Univ. of Tokyo)
pp. 271 - 278

IBISML2014-72
[Poster Presentation] Gauge Theory on Measure Space -- tau-Information Geometry --
Masaru Tanaka (Fukuoka Univ.)
pp. 279 - 286

IBISML2014-73
[Poster Presentation] Combination of LSTM and CNN on recognizing mathematical symbols
Hai Nguyen Dai, Anh Le Duc, Masaki Nakagawa (TUAT)
pp. 287 - 292

IBISML2014-74
[Poster Presentation] The method of clustering data on polar codinates
Yuji Fujjiwara (INCT), Kazuhisa Fujita (INCT/UEC)
pp. 293 - 297

IBISML2014-75
[Poster Presentation] Tensor Regression and Classification with Latent and Scaled Latent Norms
Kishan Wimalawarne (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo), Ryota Tomioka (TTIC)
pp. 299 - 306

IBISML2014-76
[Poster Presentation] Basis functions for fast learning of log-linear models
Kazuya Takabatake, Shotaro Akaho (AIST)
pp. 307 - 312

IBISML2014-77
[Poster Presentation] Infinitely exchangeable rectangular partitioning
Masahiro Nakano, Katsuhiko Ishiguro, Akisato Kimura, Takeshi Yamada, Naonori Ueda (NTT)
pp. 313 - 320

IBISML2014-78
[Poster Presentation] Enumerating all the optimal solutions of multiple travelling salesman problem by using simpath algorithm
Masashi Ogawa, Masato Inoue (Waseda Univ.)
pp. 321 - 328

IBISML2014-79
[Poster Presentation] Direct Estimation of the Derivative of Quadratic Mutual Information with Application in Sufficient Dimension Reduction
Voot Tangkaratt (Tokyo Inst. of Tech.), Hiroaki Sasaki, Masashi Sugiyama (Univ. of Tokyo)
pp. 329 - 336

IBISML2014-80
[Poster Presentation] Symbolization approach for large-scale driving corpus
Takashi Bando, Kazuhito Takenaka, Masataka Mori (DENSO), Tadahiro Taniguchi (Ritsumeikan Univ.), Chiyomi Miyajima, Kazuya Takeda (Nagoya Univ.)
pp. 337 - 343

IBISML2014-81
[Poster Presentation] Nonlinear Regression Using Deep Learning
Wataru Uchiyama, Toshiyuki Tanaka (Kyoto Univ.)
pp. 345 - 349

IBISML2014-82
[Poster Presentation] Performance analysis of auto-associative neural networks on diluted modular networks
Toshiyuki Yamane (TRL), Gouhei Tanaka (Univ. Tokyo), Daiju Nakano (TRL), Ryosho Nakane (Univ. Tokyo), Yasunao Katayama (TRL)
pp. 351 - 356

IBISML2014-83
[Poster Presentation] A Note on Attack Against Nonlinear Combiner Generator Using Sum-Product Algorithm
Kota Kubo, Shota Saito, Akira Kamatsuka, Toshiyasu Matsushima (Waseda Univ.)
pp. 357 - 364

IBISML2014-84
[Poster Presentation] Time-varying activity intensity estimation with low-rank bilinear model
Keisuke Maeda, Masamichi Shimosaka (UTokyo), Kota Tsubouchi (Yahoo!JAPAN)
pp. 365 - 371

Note: Each article is a technical report without peer review, and its polished version will be published elsewhere.