transversal study of machine learning and optimization

2011 Fiscal Year Final Research Report

• Project/Area Number: 20700251
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Single-year Grants
• Research Field: Statistical science
• Research Institution: Nagoya University
• Principal Investigator: KANAMORI Takafumi 名古屋大学, 情報科学研究科, 准教授 (60334546)
• Project Period (FY): 2008 – 2011
• Keywords: 統計的学習理論

Research Abstract

My target of this study is to provide a transversal study of machine learning and optimization theory. We apply machine learning techniques to optimization problems with noisy data. Inversely, highly developed optimization algorithms are available to conduct statistical analysis of real-world high-dimensional data. Unifying machine learning and optimization is promising for the advanced information processing.

Research Products

• [Journal Article] Worst-Case Violation of Sampled Convex Programs for Optimization with Uncertainty (2012)
  • Author(s): T. Kanamori, A. Takeda
  • Journal Title: Journal of Optimization Theory and Applications Volume: vol.152, Issue 1 Pages: 171-197
  • Peer Reviewed
• [Journal Article] F-divergence estimation and two-sample homogeneity test under semiparametric density-ratio models (2012)
  • Author(s): T. Kanamori, T. Suzuki, M. Sugiyama
  • Journal Title: IEEE Transactions on Information Theory Volume: Vol.58, Issue 2 Pages: 708-720
  • Peer Reviewed
• [Journal Article] Statistical analysis of kernel-based least-squares density-ratio estimation (2012)
  • Author(s): T. Kanamori, T. Suzuki, M. Sugiyama
  • Journal Title: Machine Learning Volume: vol.86, Issue 3 Pages: 335-367
  • Peer Reviewed
• [Journal Article] Multiscale Bagging and its Applications (2011)
  • Author(s): H. Shimodaira, T. Kanamori, M. Aoki, K. Mine
  • Journal Title: IEICE Transactions on Information and Systems Volume: Volume E94-D No.10 Pages: 1924-1932
  • Peer Reviewed
• [Journal Article] Deformation of Log-Likelihood Loss Function for Multiclass Boosting (2010)
  • Author(s): T. Kanamori
  • Journal Title: Neural Networks Volume: vol.23 Pages: 843-864
  • Peer Reviewed
• [Journal Article] Theoretical Analysis of Density Ratio Estimation (2010)
  • Author(s): T. Kanamori, T. Suzuki, M. Sugiyama
  • Journal Title: Communications and Computer Sciences Volume: vol.E93-A, no.4 Pages: 787-798
  • Peer Reviewed
• [Journal Article] A Least-squares Approach to Direct Importance Estimation (2009)
  • Author(s): T. Kanamori, S. Hido, M. Sugiyama
  • Journal Title: Journal of Machine Learning Research Volume: 10 Pages: 1391-1445
  • Peer Reviewed
• [Journal Article] Nonparametric Conditional Density Estimation Using Piecewise-Linear Path Following for Kernel Quantile Regression (2009)
  • Author(s): I. Takeuchi, K. Nomura, T. Kanamori
  • Journal Title: Neural Computation Volume: vol.21, num. 2 Pages: 533-559
  • Peer Reviewed
• [Journal Article] Robust Boosting Algorithm against Mislabelling in Multi-Class Problems (2008)
  • Author(s): T. Takenouchi, S. Eguchi, N. Murata, T. Kanamori
  • Journal Title: Neural Computation Volume: vol.20, num. 6 Pages: 1596-1630
  • Peer Reviewed
• [Presentation] Relative density-ratio estimation for robust distribution comparison (2011)
  • Author(s): M. Yamada, T. Suzuki, T. Kanamori, H. Hachiya, and M. Sugiyama
  • Organizer: Neural Information Processing Systems
  • Place of Presentation: Granada, Spain
  • Year and Date: 2011-12-13
• [Presentation] A Bregman extension of quasi-Newton updates (2010)
  • Author(s): T. Kanamori and A. Ohara
  • Organizer: Information Geometry and its Applications
  • Place of Presentation: Germany
  • Year and Date: 2010-08-02
• [Presentation] Direct density ratio estimation with dimensionality reduction SIAM (2010)
  • Author(s): Sugiyama, M., Hara, S., von Bunau, P., Suzuki, T., Kanamori, T.,& Kawanabe, M
  • Organizer: International Conference on Data Mining Columbus
  • Place of Presentation: Ohio, USA
  • Year and Date: 2010-05-29
• [Presentation] Condition Number Analysis of Kernel-based Density Ratio Estimation (2009)
  • Author(s): T. Kanamori, T. Suzuki, M. Sugiyama
  • Organizer: Numerical Mathematics in Machine Learning(NUMML2009)
  • Place of Presentation: Montreal, Canada
  • Year and Date: 2009-06-18
• [Presentation] Efficient Direct Density Ratio Estimation for Non-stationarity Adaptation and Outlier Detection (2008)
  • Author(s): T. Kanamori, M. Sugiyama, and S. Hido
  • Organizer: Neural Information Processing Systems
  • Place of Presentation: Vancouver, B. C., Canada
  • Year and Date: 2008-12-16
• [Book] Density Ratio Estimation in Machine Learning (2012)
  • Author(s): M. Sugiyama, T. Suzuki, T. Kanamori
  • Total Pages: 119-297
  • Publisher: Cambridge University Press
• [Book] パターン認識(Rで学ぶデータサイエンス5) (2009)
  • Author(s): 金森敬文,竹之内高志,村田昇
  • Total Pages: 1-153
  • Publisher: 共立出版