Researcher's Profile

While deep learning is attracting much attention these days, we are especially interested in unsupervised learning, which is one of main topics in machine learning research. An important purpose of unsupervised learning is to reveal latent structures or patterns such as clusters and low-dimensional intrinsic subspace behind the big data and to model the mechanism of the data generation.

We are also studying on algorithms of learning dynamical systems (LDS), which aims at identifying mathematical models of natural and artificial dynamical systems from observation data. Obtained models can be utilized for control and prediction of those systems. This topic is closely reated to control engineering as well to machine learning.

Furthermore, we are applying these techniques to health monitoring and anomaly detection of large-scale artificial systems such as artificial satellites and plants. The goal of our study is to make the world safer and more secure by these technologies.