Academic Employment

2019- Assistant Professor, Kyoto University

Education

2014-2018 Ph.D. program in Statistics, The Graduate University for Advanced Studies

2012-2014 MS in Biostatistics, Kitasato University

2007-2008 Ph.D. program in Mathematics, University of Southern California

2006-2007 Ph.D. program in Mathematics, University at Buffalo, SUNY

2003-2006 BS in Mathematics, University at Buffalo, SUNY

Research Interests

Algebraic Statistics

Singular Learning Theory

Information Criteria

Causal Discovery

Application of Deep Learning to Medicine (especially, computational pathology)

Awards

2017 Presentation Award of Japanese Society for Quality Control

2016 Best Presentation Award of Japanese Society of Applied Statistics

2005 Harriet F. Montague Award

Papers

Working Papers

-Imai T. Estimating real log canonical thresholds, arXiv:1906.01341

-Imai T. On the overestimation of widely applicable Bayesian information criterion, arXiv:1908.10572

-Imai T. Bayesian estimation of singular Bayesian information criterion

-Imai T. Analysis for the constant order term of the asymptotic expansion of the marginal likelihood for singular models

Discussion

Imai, T., Kuroki, M. (2017) Discussion of A Bayesian information criterion for singular models, Journal of the Royal Statistical Society. Series B (Methodological), 79 (2), p363-364

Talks

2018 Estimating real log canonical thresholds of statistical models, Summer School in Algebraic Statistics in Tromsø

2017 An optimal temperature for accurate model evidence evaluation, Algebraic and Geometric Methods in Statistics at Institute of Mathematical Statistics