Hello! I’m Hiro.
Hirofumi Shiba is a Ph.D. student jointly supervised by Kengo Kamatani and Keisuke Yano at the Institute of Statistical Mathematics (ISM), Tokyo, Japan.
My research focuses around stochastic algorithms and numerical computation. By making them more efficient, scalable, and automated, I aim to broaden the applications of statistics and machine learning.
Before pursuing my Ph.D., I studied Mathematics at the University of Tokyo.
Interests
- Sampling via Transport (Schrödinger Bridge, Interacting Particle Method, etc.)
- Monte Carlo Computation (SMC, MCMC, PDMP, etc.)
- Bayesian Machine Learning (Gaussian Process, Nonparametrics, etc.)
- Bayesian Modeling (Political Science, Biostatistics, Planetary Science, etc.)
- Data-driven Science (Diffusion Map, Trajectory Learning, etc.)
Research
My research interests revolve around developing and analyzing sampling algorithms, by leveraging insights from their continuous-time limit dynamics. From this perspective, algorithms reveal their intrinsic properties, and my work aims to unify the understanding of various algorithms under a common framework.
Additionally, I engage in Bayesian data analysis in collaboration with healthcare and engineering companies.
Education
Ph.D. in Statistical Science, 2028 (expected)
Graduate University for Advanced Studies, SOKENDAI, Japan
B.A. in Mathematics, 2023
University of Tokyo, Japan
Experience
Cooperative Researcher
RCAST, the University of Tokyo
Economic Security Research Program
Consultant
PreMedica Inc., Tokyo, Japan