Centers & Labs

RIKEN Center for Advanced Intelligence Project

Computational Learning Theory Team

Team Leader: Kohei Hatano (D.Sci.)
Kohei  Hatano(D.Sci.)

We try to formulate and solve various problems in machine learning from a theoretical computer science perspective. One of our primary research topics is online decision making problems, where the (possibly adversarial) environment and the player interact iteratively. Our goal is to clarify the limits of the player’s strategies for various online decision problems under continuous/discrete constraints and to develop robust and efficient strategies based on theoretical analyses. We also keep connections with other different areas and investigate new applications of machine learning techniques.

Main Research Field

Computer Science

Research Subjects

  • Online Prediction
  • Machine Learning
  • Applications of machine learning to other disciplines

Selected Publications

Papers with an asterisk(*) are based on research conducted outside of RIKEN.
  1. *Ailon, N., Hatano, K., and Takimoto, E.:
    “Bandit Online Optimization Over the Permutahedron”
    Theoretical Computer Science, vol. 650, pp.92-108, (2016).
  2. *Nakazono, T., Moridomi, K., Hatano, K., and Takimoto, E.:
    “A Combinatorial Metrical Task System Problem under the Uniform Metric”
    Proceedings of the 26th International Conference on Algorithmic Learning Theory (ALT 2016), LNCS vol. 9925, pp. 276-287, (2016).
  3. *Matsumoto, I., Hatano, K., and Takimoto, E.:
    “Online Density Estimation of Bradley-Terry Models”
    Proceedings of the 28th Conference on Learning Theory (COLT 2015), JMLR W&CP vol.40, pp.1343–1359, (2015).
  4. *Fujita, T., Hatano, K., Kijima, S., and Takimoto, E.:
    “Online Linear Optimization for Job Scheduling under Precedence Constraints”
    Proceedings of 26th International Conference on Algorithmic Learning Theory (ALT 2015), LNCS vol.6331, pp.345–359, (2015).
  5. *Moridomi, K., Hatano, K., Takimoto, E., and Tsuda, K.:
    “Online matrix prediction for sparse loss matrices”
    Proceedings of the Sixth Asian Conference on Machine Learning (ACML 2014), JMLR W&CP vol.39, pp.250–265, (2015).
  6. *Fujita, T., Hatano, K., and Takimoto, E.:
    “Combinatorial Online Prediction via Metarounding,”
    Proceedings of 24th Annual Conference on Algorithmic Learning Theory (ALT 2013), LNCS vol.8139, pp.68–82, (2013).
  7. *Suehiro, D., Hatano, H., Kijima, S., Takimoto, E., and Nagano, K.:
    “Online Prediction under Submodular Constraints”
    Proceedings of 23th Annual Conference on Algorithmic Learning Theory (ALT 2012), LNCS vol.7568, pp.260–274, (2012).
  8. *Anan, Y., Hatano, K., Bannai, H., and Takeda, M.:
    “Music Genre Classification using Similarity Functions”
    Proceedings of the 12th International Society for Music Information Retrieval Conference (ISMIR 2011), pp.693–698, (2011).
  9. *Yasutake, S., Hatano, K., Kijima, S., Takimoto, E., and Takeda, M.:
    “Online Linear Optimization over Permutations”
    Proceedings of the 22nd International Symposium on Algorithms and Computation (ISAAC 2011), LNCS vol.7074, pp.534–543, (2011).
  10. *Ishibashi, K., Hatano, K., and Takeda, M.:
    “Online Learning of Maximum p-Norm Margin Classifiers with Bias”
    Proceedings of the 21st Annual Conference of Learning Theory (COLT 2008), pp.69–80, (2008).

Contact information

R&D Division, Library, Kyushu University,
6-10-1 Hakozaki, Higashi-ku,
Fukuoka, Japan

Email: hatano [at] inf.kyushu-u.ac.jp

Related links