| |
- |
|
Mon, Jun 14 AM Opening
09:30 - 09:35 |
| |
- |
Greetings from IBISML steering committee |
Mon, Jun 14 AM Special session 1
Chair: Tsuyoshi Ide (IBM)
09:35 - 12:10 |
| (1) |
09:35-10:25 |
[Invited Talk]
Advances in Statistical Machine Learning
-- An Approach based on Probability Density Ratios -- |
Masashi Sugiyama (Tokyo Inst. of Tech.) |
| (2) |
10:25-11:15 |
[Invited Talk]
All Pairs Similarity Search by Multiple Sorting |
Koji Tsuda (AIST) |
| |
11:15-11:20 |
Break ( 5 min. ) |
| (3) |
11:20-12:10 |
[Invited Talk]
Relational Data Mining on Causal Relations Between Variables |
Takashi Washio (Osaka Univ.) |
| |
12:10-13:30 |
Lunch Break ( 80 min. ) |
Mon, Jun 14 PM Special session 2
Chair: Kenji Yamanishi (U. Tokyo)
13:30 - 15:20 |
| (1) |
13:30-14:20 |
[Invited Talk]
Introduction and Recent Advance of Algebraic Geometry and Learning Theory |
Sumio Watanabe (Tokyo Inst. of Tech.) |
| |
14:20-14:30 |
Break ( 10 min. ) |
| (2) |
14:30-15:20 |
[Invited Talk]
Large-Scale String Processing: Theory and Practice |
Daisuke Okanohara (Preferred Infrastructure Icn.) |
| |
15:20-15:30 |
Break ( 10 min. ) |
Mon, Jun 14 PM Contributed talks (new approaches)
Chair: Toshihiro Kamishima (AIST)
15:30 - 16:15 |
| (1) |
15:30-15:45 |
A Study on the Exact Nonlinear Regularization Path for L2 Loss Support Vector Machines |
Masayuki Karasuyama, Ichiro Takeuchi (Nagoya Inst. of Tech.) |
| (2) |
15:45-16:00 |
Hierarchical Pitman-Yor Topic Model |
Issei Sato, Hiroshi Nakagawa (Univ. of Tokyo.) |
| (3) |
16:00-16:15 |
Orthogonalization and thresholding method for a nonparametric regression problem |
Katsuyuki Hagiwara (Mie Univ.) |
| |
16:15-16:25 |
Break ( 10 min. ) |
Mon, Jun 14 PM Contributed talks (combining models and data)
Chair: Hisashi Kashima (U. Tokyo)
16:25 - 17:40 |
| (1) |
16:25-16:40 |
Privacy Preservation in Online Prediction |
Jun Sakuma (Univ. of Tsukuba/JST), Hiromi Arai (U. Tsukuba) |
| (2) |
16:40-16:55 |
Collaborative Filtering with A Bayesian Hierarchical Model |
Hideki Asoh (AIST) |
| (3) |
16:55-17:10 |
Customized Pricing Recommender System
-- Simple Implementation and Preliminary Experiments -- |
Toshihiro Kamishima, Shotaro Akaho (AIST), Jun Sakuma (Univ. of Tsukuba/JST) |
| (4) |
17:10-17:25 |
Regularized Random Forest Method for Survival Analysis |
Toshio Shimokawa (Yamanashi Univ.), Mitsuhiro Tsuji (Kansai Univ.) |
| (5) |
17:25-17:40 |
Multifaceted Data Mining with Influence Diffusion Model |
Naohiro Matsumura (Osaka Univ.) |
| |
18:30-20:30 |
Banquet (Place: Lever son Verre Hongo http://www.madoi-co.com/food/leversonverre/hongo.html ) ( 120 min. ) |
Tue, Jun 15 AM Special session 3
Chair: Tsuyoshi Ide (IBM)
09:30 - 12:10 |
| (1) |
09:30-10:20 |
[Invited Talk]
Statistical Machine Learning Based on Nonparametric Bayesian Models |
Takaki Makino (Univ. of Tokyo.) |
| |
10:20-10:25 |
Break ( 5 min. ) |
| (2) |
10:25-11:15 |
[Invited Talk]
Statistical testing with large multiplicity |
Shigeyuki Oba (Kyoto Univ./JST) |
| |
11:15-11:20 |
Break ( 5 min. ) |
| (3) |
11:20-12:10 |
[Invited Talk]
Machine Learning in Object Recognition |
Keiji Yanai (Univ. of Electro-Comm.) |
| |
12:10-13:30 |
Lunch break ( 80 min. ) |
| |
- |
Announcement (Shigeyuki Oba, Daichi Mochihashi) |
Tue, Jun 15 PM Special session 4
Chair: Kenji Yamanishi (U. Tokyo)
13:30 - 14:20 |
| (1) |
13:30-14:20 |
[Invited Talk]
Recent Developments in Markov-chain Monte Carlo Method |
Koji Hukushima (Univ. of Tokyo.) |
| |
14:20-14:30 |
Break ( 10 min. ) |
Tue, Jun 15 PM Contributed talks (learning for physics)
Chair: Daichi Mochihashi (NTT)
14:30 - 15:00 |
| (1) |
14:30-14:45 |
Inference of Multiple Absorbtion Bands for Reflectance Spectra Using Exchange Monte Carlo Method |
Kenji Nagata, Seiji Sugita (Univ. of Tokyo), Masato Okada (Univ. of Tokyo/RIKEN) |
| (2) |
14:45-15:00 |
Efficient estimation of phase response curves using a data transformation |
Ken Nakae (SOKENDAI) |
| |
15:00-15:10 |
Break ( 10 min. ) |
Tue, Jun 15 PM Contributed talks (reinforcement learning)
Chair: Shigeyuki Oba (Kyoto U.)
15:10 - 15:40 |
| (1) |
15:10-15:25 |
Generalization of TD-learning from a Semiparametric Statistical Viewpoint |
Tsuyoshi Ueno, Shin-ichi Maeda (Kyoto Univ.), Motoaki Kawanabe (Fraunhofer First), Shin Ishii (Kyoto Univ.) |
| (2) |
15:25-15:40 |
New Feature Selection Method for Reinforcement Learning
-- Conditional Mutual Information Reveals Implicit State-Reward Dependency -- |
Hirotaka Hachiya, Masashi Sugiyama (Tokyo Insst. of Tech.) |
| |
15:40-15:50 |
Break ( 10 min. ) |
Tue, Jun 15 PM Contributed talks (probabilistic networks)
Chair: Takashi Washio (Osaka U.)
15:50 - 16:35 |
| (1) |
15:50-16:05 |
Dependence Minimizing Regression with Model Selection for Non-Linear Causal Inference under Non-Gaussian Noise |
Makoto Yamada, Masashi Sugiyama (Tokyo Inst. of Tech.) |
| (2) |
16:05-16:20 |
An EM algorithm on binary decision diagrams with order encoding for propositional logic-based probabilistic models |
Masakazu Ishihata, Yoshitaka Kameya, Taisuke Sato (Tokyo Inst. of Tech.), Shin-ichi Minato (Hokkaido Univ) |
| (3) |
16:20-16:35 |
Computational Customer Modeling by Using the Bayesian Network and the Double-latent Class Model in a Retail Service |
Tsukasa Ishigaki, Takeshi Takenaka, Yoichi Motomura (AIST) |
| |
16:35-16:45 |
Break ( 10 min. ) |
Tue, Jun 15 PM Contributed talks (model selection)
Chair: Masashi Sugiyama (Tokyo Tech.)
16:45 - 17:45 |
| (1) |
16:45-17:00 |
A linear time algorithm for sequential dynamic model selection |
Eiichi Sakurai, Kenji Yamanishi (Univ. of Tokyo.) |
| (2) |
17:00-17:15 |
The Coding Divergence for Measuring the Complexity of Separating Two Sets |
Mahito Sugiyama, Akihiro Yamamoto (Kyoto Univ.) |
| (3) |
17:15-17:30 |
Graph Clustering based on Normalized Maximum Likelihood Coding |
So Hirai, Ryota Tomioka, Kenji Yamanishi (Univ. of Tokyo) |
| (4) |
17:30-17:45 |
Biases of error rate by selection of models or algorithms and estimating the true error rate |
Issei Kurahashi, Yasuo Ohashi (Univ. of Tokyo.) |
Tue, Jun 15 PM Closing
17:45 - 17:50 |
| |
- |
Closing remarks and introduction to the next workshop |