The Ph.D. thesis:

The full list of publications is here.

The followings are selected papers.

Theoretical Analysis of Bayesian Statistics

  1. Keisuke Yamazaki and Sumio Watanabe, "Singularities in mixture models and upper bounds of stochastic complexity", Neural Networks, 16, 1029-1038, 2003.
  2. Keisuke Yamazaki and Sumio Wataanbe, "Algebraic geometry and stochastic complexity of hidden Markov models", Neurocomputing, 69, 62-84, 2005.
  3. Keisuke Yamazaki, Motoaki Kawanabe, Sumio Watanabe, Masashi Sugiyama and Klaus-Robert Mueller, "Asymptotic Bayesian generalization error when training and test distributions are different", Proc. of ICML 2007, 1079-1086, 2007.
  4. Keisuke Yamazaki and Samuel Kaski, "An Analysis of Generalization Error in Relevant Subtask Learning", Proc. of ICONIP 2008, 629-637, 2008.
  5. Keisuke Yamazaki, Miki Aoyagi, Sumio Watanabe, "Asymptotic Analysis of Bayesian Generalization Error with Newton Diagram", Neural Networks, 23, 35-43, 2010.

Accuracy of Unsupervised Learning

  1. Keisuke Yamazaki, "An Asymptotic Analysis of Bayesian State Estimation in Hidden Markov Models", Proc. of Machine Learning for Signal Processing, 2011.
  2. Keisuke Yamazaki, "Asymptotic Accuracy of Distribution-Based Estimation of Latent Variables", Journal of Machine Learning Research, 15, 3541-3562, 2014.
  3. Keisuke Yamazaki, "Asymptotic Accuracy of Bayes Estimation for Latent Variables with Redundancy", Machine Learning, to appear.
  4. Keisuke Yamazaki, "Accuracy of Latent-Variable Estimation in Bayesian Semi-Supervised Learning", Neural Networks, 69, 1-10, 2015.

Application to Traffic Flow Analysis

  1. Koichi Kobayashi and Keisuke Yamazaki, "Parameter Estimation Accuracy and Active Learning in the Zero-Range Process", Proc. Soft Computing and Pattern Recognition, 48-51, 2012.
  2. Fumito Nakamura and Keisuke Yamazaki, ”Two Statistical Methods for Grouping Vehicles in Traffic Flow Based on Probabilistic Cellular Automata”, Proc. of SCIS & ISIS ,2014.