Research Director

Kenji Nagata

National Institute of Advanced Industrial Science and Technology (AIST)
Senior Researcher

Outline

For the statistical stochastic models with hierarchical structure such as Deep networks, Gaussian mixture models and hidden Markov models, algebraic geometry have contributed to the performance elucidation of statistical learning. Based on this knowledge, this study develops a new technology of data-driven approach with Bayesian estimation. Moreover, this study promotes an academic area to be called computational algebraic geometry with the knowledge of computer science and statistics.