Welcome to Takuya’s Homepage
Bio
- 2024-present. PhD student in Econometrics and Statistics at The University of Chicago, Booth School of Business. Advisor: Tengyuan Liang.
- 2022-2024. PhD student in Statistics at Rutgers University. Advisor: Pierre C. Bellec.
- 2018-2022. BE in Mathematical Engineering (=Math & CS), The University of Tokyo.
- 2021-2022. Student researcher at RIKEN Center for Brain Science.
Research
I have broad interests in mathematical theory and algorithms. In recent years, my research has focused on: (1) Mean-field high dimensional statistics (2) High-dimensional probability, and (3) Combinatorial stochastic processes. More recently, I have been studying optimal transport and learning theory, with a particular emphasis on generative model and reinforcement learning.
Preprints
Parameter estimation of Gibbs-partitions with quasi-likelihood: stable martingale CLT and α-diversity
Takuya Koriyama, in preparation.Phase transitions for the existence of unregularized M-estimators in single index models
Pierre C Bellec and Takuya Koriyama (alphabetical order)
arXiv:2501.03163, 2025, submitted.Denoising Diffusions with Optimal Transport: Localization, Curvature, and Multi-Scale Complexity
Tengyuan Liang, Kulunu Dharmakeerthi, and Takuya Koriyama
arXiv:2411.01629, 2024, submitted.Precise Asymptotics of Bagging Regularized M-estimators
Takuya Koriyama, Pratik Patil, Jin-Hong Du, Kai Tan, and Pierre C. Bellec
arXiv:2409.15252, 2024, submitted. [github]Asymptotics of resampling without replacement in robust and logistic regression
Pierre C. Bellec and Takuya Koriyama (alphabetical order)
arXiv:2404.02070, 2024, submitted.Existence of solutions to the nonlinear equations characterizing the precise error of M-estimators
Pierre C. Bellec and Takuya Koriyama (alphabetical order)
arXiv:2312.13254, 2023, submitted. [slide]Asymptotic analysis of parameter estimation for the Ewens–Pitman partition
Takuya Koriyama, Takeru Matsuda and Fumiyasu Komaki
Minor revision at Journal of Applied Probability. arXiv:2207.01949, 2022. [slide].
Publications
Error estimation and adaptive tuning for unregularized robust M-estimator
Pierre C. Bellec and Takuya Koriyama (alphabetical order)
Journal of Machine Learning Research 26 (16), 1-40, 2025.Corrected generalized cross-validation for finite ensembles of penalized estimators
Pierre C. Bellec, Jin-Hong Du, Takuya Koriyama, Pratik Patil and Kai Tan (alphabetical order)
Journal of the Royal Statistical Society Series B: Statistical Methodology, qkae092, 2024. [journal], [slide], [github].Fully Data-driven Normalized and Exponentiated Kernel Density Estimator with Hyvärinen Score
Shunsuke Imai, Takuya Koriyama, Shouto Yonekura, Shonosuke Sugasawa and Yoshihiko Nishiyama
Journal of Business & Economic Statistics 43 (1), 110-121, 2025. [journal]
Talk
- IMS International Conference on Statistics and Data Science, Nice, December 2024.
- Workshop Honoring Andrew Barron: Forty Years at the Interplay of Information Theory, Probability and Statistical Learning, Yale University, April 2024.
- IMS Asia Pacific Rim Meeting, Melbourne, January 2024.
- Seminar talk, Graduate school of Economics, the University of Tokyo, February 2023.
- IMS International Conference on Statistics and Data Science, Florence, December 2022.
- Workshop on random partition of integers, the University of Tokyo, June 2022.
- Japan Statistical Society Meeting, Keio University, March 2022.
Teaching (PhD level)
- TA: Theory of Probability, fall 2023.
- TA: Stochastic process, spring 2024.
- Instructor: Probability and Stochastic Processes, summer 2024.
Award
- IMS Hannan Graduate Student Travel Award, 2025.
- Travel award for Yale FDS workshop, 2024.
- Best Ph.D. Qualifying Exam Performance, Department of Statistics, Rutgers University, 2023.
- Student Travel Award, IMS International Conference on Statistics and Data Science, 2022.
- Best Presentation Award and Japan Statistical Society Certificate Director’s Award, 16th Japan Statistical Society Spring Meeting, 2022.