データマイニングと統計数理研究会（第 10 回）目次

(1)

人工知能学会研究会資料 SIG-DMSM-A901

データマイニングと統計数理研究会（第

10 回）

テーマ：「データマイニングと統計数理」

The International Workshop on Data Mining and Statistical Science

(DMSS2009) と共催

日時：2009 年 7 月 7 日（火）～8 日（水）会場：京大会館

(2)

Program

July 7

[13:20-13:30] Opening Remark

Session 1

[13:30-15:00]

1. Efficient Exploration through Active Learning

for Value Function Approximation in Reinforcement Learning Takayuki Akiyama, Hirotaka Hachiya, and

Masashi Sugiyama (Tokyo Institute of Technology) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥001 2. Visual Inspection of Precision Instruments by Least-Squares Outlier Detection

Masafumi Takimoto, Masakazu Matsugu (Canon), and

Masashi Sugiyama (Tokyo Institute of Technology) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥022 3. Extracting Phases of Financial Markets

Teruko Takada (Osaka City University) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥027 [15:00-15:15] Break

Session 2

[15:15-16:15]

Recent Topics on BDDs/ZDDs for Data Mining and Knowledge Discovery (Invited Talk) Shin-ichi Minato (Hokkaido University)

[16:15-16:30] Break

Session 3

[16:30-17:30]

4. Dimensionality Reduction for Density Ratio Estimation in High-dimensional Spaces Masashi Sugiyama (Tokyo Institute of Technology), Motoaki Kawanabe

(Fraunhofer Institute FIRST), and Pui Ling Chui (Tokyo Institute of Technology) ‥031 5. Sufficient Dimension Reduction via Squared-loss Mutual Information Estimation

Taiji Suzuki (University of Tokyo) and

(3)

July 8

Session 4

[10:00-11:30]

6. Identification of an Exogenous Variable

in a Linear non-Gaussian Structural Equation Model

Shohei Shimizu (Osaka University), Aapo Hyvarinen (University of Helsinki), Yoshinobu Kawahara (Tokyo Institute of Technology), and

TakashiWashio (Osaka University) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥078 7. Maximum Likelihood Estimation for Failed-Link Detection in Communication Networks

Shohei Hido (Kyoto University, IBM Research) and

Yutaka Takahashi (Kyoto University) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥087 8. Gaussian Mixture Models and VC Dimensions

Yohji Akama (Tohoku University) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥094 [11:30-13:30] Lunch

Session 5

[13:30-14:30] Invited Talk

Katsutoshi Yada (Kansai University)

[14:30-14:45] Lunch

Session 6

[14:45-16:15]

9. Monolithic and Partial Compilation Methods for Probabilistic Inference of Bayesian Networks using ZBDDs

Daisuke Tokoro, Kiyoharu Hamaguchi, Toshinobu Kashiwabara (Osaka University), and Shin-ichi Minato (Hokkaido University) ‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥‥098 10. Pattern Discovery from a Single Graph with Quantitative Itemsets

Yuuki Miyoshi, Tomonobu Ozaki, and Takenao Ohkawa (Kobe University) ‥‥‥‥104 11. Mining Frequent Patterns from Linear Graphs

Yasuo Tabei, Daisuke Okanohara (University of Tokyo), and Koji Tsuda(AIST) ‥‥‥111 [16:15-16:20] Closing Remark