Mathieu Blondel

Welcome to my little corner on the Internet! I'm a French PhD student, currently studying and living in Kobe, Japan.

You can reach me at: mathieu !AT! mblondel *DOT* org

Quick links:

• Blog
• Tweets
• Resume
• Github, Gist

Research Interest

My main research interest is Machine Learning and its applications.

Publications

Journal Papers

• Scikit-learn: Machine Learning in Python. [PDF, www]
  Fabian Pedregosa, Gaël Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, Jake Vanderplas, Alexandre Passos, David Cournapeau, Matthieu Brucher, Matthieu Perrot, Édouard Duchesnay.
  Journal of Machine Learning Research volume 12, pp. 2825−2830, 2011.
• Literature-based gene function annotation with kernels.
  Mathieu Blondel, Kazuhiro Seki, and Kuniaki Uehara.
  IPSJ Transactions on Mathematical Modeling and its Applications volume 4, pp. 49-58, 2011.

International Conference Papers (Peer reviewed)

• Application of Semantic Kernels to Literature-Based Gene Function Annotation.
  Mathieu Blondel, Kazuhiro Seki, and Kuniaki Uehara.
  In Proceedings of the 14th International Conference on Discovery Science (DS 2011), October 2011.
• Tackling Class Imbalance and Data Scarcity in Literature-Based Gene Function Annotation. [PDF]
  Mathieu Blondel, Kazuhiro Seki, and Kuniaki Uehara.
  In Proceedings of the 31th annual international ACM SIGIR conference on research and development in information retrieval (SIGIR 2011), July 2011.
  (poster, acceptance rate: 89/274 = 32%)
• Unsupervised Learning of Stroke Tagger for Online Kanji Handwriting Recognition. [PDF]
  Mathieu Blondel, Kazuhiro Seki, and Kuniaki Uehara.
  In Proceedings of the 20th International Conference on Pattern Recognition (ICPR 2010), August 2010.

Software

scikit-learn is an easy-to-use and efficient machine learning library for the Python scientific ecosystem. (Committer and active developer)

Tegaki is a handwriting recognition project for Chinese and Japanese characters. Tegaki is multi-plaform and supports several recognition engines. (Main developer)

Fantasdic is a dictionary application which allows you to look up words in a variety of dictionary sources. It is primarily targetting the GNOME desktop but it works on other platforms too. nihongobenkyo, a server containing Japanese dictionaries that I maintain, can be used with Fantasdic. (Main developer)

Academic Activities

Program Committee member of the 7th Asian Information Retrieval Societies Conference (AIRS 2011).

Scholarships

Japanese Government Monbukagakusho Research Scholarship, September 2009 to March 2013.