Jun-ichiro Hirayama

Engilsh/Japanese

Jun-ichiro HIRAYAMA

I have moved to ATR. This page is no longer maintained. Please visit my new home page.

Postdoctral fellow,
Graduate School of Informatics,
Kyoto University

Publications

Journal Papers

J. Hirayama, M. Nakatomi, T. Takenouchi and S. Ishii.
Collaborative prediction by multiple Bayesian networks and its application to printer usage modeling.
Behaviormetrika, 35(2), 99-114, 2008.
J. Hirayama, M. Nakatomi, T. Takenouchi and S. Ishii.
α-Bayesian collaboration of multiple predictors and its applications to hybrid recommendation and user modeling.
Neural Information Processing - Letters and Reviews, 12(1-3), 11-20, 2008.
S. Osaga, J. Hirayama, T. Takenouchi and S. Ishii.
A probabilistic modeling of MOSAIC learning.
Artificial Life and Robotics, 12, 167-171, 2008.
J. Hirayama, S. Maeda and S. Ishii.
Markov and semi-Markov switching of source appearances for non-stationary independent component analysis.
IEEE Transactions on Neural Networks, 18(5), 1326-1342, 2007.
J. Hirayama, J. Yoshimoto and S. Ishii.
Balancing plasticity and stability of on-line learning based on hierarchical Bayesian adaptation of forgetting factors.
Neurocomputing, 69(16-18), 1954-1961, 2006.
J. Hirayama, J. Yoshimoto and S. Ishii.
Bayesian representation learning in the cortex regulated by acetylcholine.
Neural Networks, 17(10), 1391-1400, 2004.

Conference Proceedings

J. Hirayama and A. Hyvärinen.
Structural equations and divisive normalization for energy-dependent component analysis.
NIPS2011. (to appear).
M. U. Gutmann and J. Hirayama.
Bregman divergence as general framework to estimate unnormalized statistical models.
Proc. Conf. on Uncertainty in Artificial Intelligence (UAI2011), pp.283-290, 2011.
J. Hirayama, A. Hyvärinen and S. Ishii.
Sparse and low-rank estimation of time-varying Markov networks with alternating direction method of multipliers.
International Conference on Neural Information Processing (ICONIP'10), 2010.
Lecture Notes in Computer Science, 6443, pp.371-379, 2010.
K. Hayashi, J. Hirayama and S. Ishii.
Dynamic Exponential Family Matrix Factorization.
The 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD'09), 2009.
Lecture Notes in Computer Science, 5476, pp.452-462, 2009.
A. Asahina, J. Hirayama and S. Ishii.
Interpreting dopamine activities in stochastic reward tasks.
International Conference on Neural Information Processing (ICONIP'08), 2008. *Best APNNA student paper award
Lecture Notes in Computer Science, 5506, pp.361-368, 2009.
J. Hirayama and S. Ishii.
A closed-form estimator of fully visible Boltzmann machines.
International Conference on Neural Information Processing (ICONIP'08), 2008.
Lecture Notes in Computer Science, 5507, pp.951-959, 2009.
A. Ishihara, J. Hirayama, T. Takenouchi and S. Ishii.
A unified approach to collaborative and feature-based recommendation based on probabilistic latent semantic models.
International Symposium on Artificial Life and Robotics (AROB'08), GS17-2, 2008.
J. Hirayama, M. Nakatomi, T. Takenouchi and S. Ishii.
Bayesian collaborative predictors for general user modeling tasks.
International Conference on Neural Information Processing (ICONIP'07), 2007.
Lecture Notes in Computer Science, 4984, pp.742-751, 2008.
S. Osaga, J. Hirayama, T. Takenouchi and S. Ishii.
A probabilistic model of MOSAIC.
The First IEEE Symposium on Foundations of Computational Intelligence (FOCI'07), pp.41-46, 2007.
S. Osaga, J. Hirayama, T. Takenouchi and S. Ishii.
A probabilistic modeling of MOSAIC learning.
Artificial Life and Robotics (AROB'07), GS16-3, 2007.
J. Hirayama, S. Maeda and S. Ishii
A Bayesian approach to blind source separation with variable number of sources.
Artificial Life and Robotics (AROB'06), GS19-6, 2006. *Young author award
J. Hirayama, S. Maeda and S. Ishii.
Bayesian noisy ICA for source switching environments.
IEEE Workshop on Statistical Signal Processing (SSP'05), 232, 2005.
J. Hirayama, J. Yoshimoto and S. Ishii.
Cortical representation learning regulated by acetylcholine.
Brain Inspired Cognitive Systems (BICS'04), ICESS.3, 2004.

Technical Reports

J. Hirayama, J. Yoshimoto and S. Ishii.
Balancing plasticity and stability of on-line learning based on hierarchical Bayesian adaptation of forgetting factors.
NAIST Technical Report, NAIST-IS-TR2005008, Nara Institute of Science and Technology, 2005.

Japanese page

Biography

1979.10	Born in Fukuoka , Kyu-shu, Japan
1992.4-1995.3	Kurume University Attached (Fusetsu) Junior High School.
1995.4-1998.3	Go on to the same Fusetsu High School.
1998.4-2002.3	Dept. of Materials Science and Engineering, Kyoto University.
2002.4-2007.3	Graduate School of Information Science, Nara Institute of Science and Technology
2006.4-2008.3	Research Fellow of the Japan Society for the Promotion of Science (DC2), Graduate School of Information Science, Nara Institute of Science and Technology.
2008.4-2009.3	Postdoctoral Fellow, Graduate School of Informatics, Kyoto University.
2009.4(-2012.3)	Research Fellow of the Japan Society for the Promotion of Science (PD), Graduate School of Informatics, Kyoto University.
2010.1-2011.7	Visiting researcher at Neuroinformatics Group, Helsinki Institute of Information Technology.

Jun-ichiro HIRAYAMA
hirayama_at_robot.kuass.kyoto-u.ac.jp

Mechanical Systems Control Lab.,
Graduate School of Informatics,
Kyoto University.