Katsuyuki Hagiwara
Faculty of Education, Mie University,
1515 Kamihama, Tsu, 514-8507 JAPAN
E-MAIL : hagi@edu.mie-u.ac.jp

Japanese page

---

Biography

He received a PhD in Engineering from the Toyohashi University of Technology in 1995.
He joined the Faculty of Engineering, Mie University in 1995.
In 2003, he joined the Nagoya Institute of Technology.
In 2004, he then joined the Faculty of Education, Mie University.

Research Interests

• statistical theory of multi-layer perceptron
• regularization and model selection
• statistical signal processing

Recent publications

• Hagiwara, K. and Fukumizu, K.,
  ``Relation between weight size and degree of over-fitting in neural network regression'',
  Neural Networks, 21, pp.48-58, 2008.
• Hagiwara, K., ``Orthogonal Shrinkage Methods for Nonparametric Regression under Gaussian Noise'',
  Lecture Notes in Computer Science 4984 -- 14th International Conference, ICONIP 2007, Ishikawa,M., Doya,K., Miyamoto,H., and Yamakawa,T. Eds., Kitakyushu, Japan, November 13-16, 2007, Part I, pp.537-546.
• Hagiwara, K. and Ishitani, H., ``Estimation of the expected prediction error of orthogonal regression with variable components'', 2007 Hawaii International Conference on Statistics, Mathematics and Related Fields, Honolulu, Hawaii, pp.427-431, 2007.
• Hagiwara, K. and Ishitani, H., ``On the expected prediction error of orthogonal regression with variable components'',
  IEICE Trans. Fundamentals, E89-A, 12, 3699-3709, 2006.
• Hagiwara, K., ``A consistent model selection for orthogonal regression under component--wise shrinkage'',
  Journal of Statistical Planning and Inference,136,1181-1195, 2006
• Hagiwara, K. and Fukumizu, K, `` A probabilistic upper bound for the degree of over-fitting to noise in neural network regression'',
  Research Memorandum No.953, Institute of Statistical Mathematics. 2005.
• Hagiwara, K. and Fukumizu, K, ``Over-fitting behavior of Gaussian unit under Gaussian noise'',
  Proc. of IJCNN2004, 2, 997-1002, 2004.
• Fukumizu, K and Hagiwara, K., ``A general upper bound of likelihood ratio for regression'',
  Research Memorandum No.887, Institute of Statistical Mathematics. 2003.
• Hagiwara, K., ``Model selection with componentwise shrinkage in orthogonal regression'',
  IEICE Trans. Fundamentals, E86-A, 7, 1749-1758, 2003.
• Fukumizu, K and Hagiwara, K., ``A general upper bound of likelihood ratio in binary regression'',
  Research Memorandum No.858, Institute of Statistical Mathematics. 2002.
• Hagiwara, K., ``Regularization learning, early stopping and biased estimator'',
  Neurocomputing, 48, 937-955, 2002.
• Hagiwara, K., ``On the problem in model selection of neural network regression in overrealizable scenario'',
  Neural Computation, 14, pp.1979-2002, 2002.
• Hagiwara, K., Hayasaka, T., Toda, N., Usui, S. and Kuno, K.,
  ``Upper bound of the expected training error of neural network regression for a Gaussian noise sequence'',
  Neural Networks, 14, pp.1419-1429, 2001.

---