Shuhei MANO

E-mail: smano at ism.ac.jp

The Institute of Statistical Mathematics
10-3 Midori-cho, Tachikawa, Tokyo 190-8562, Japan

More information is available on my page in Researchmap.

Research Interests

Modelling underlies human recognition. It induces mathematical structure, and this structure gives algorithms for data-based inference and computation.

I work in statistics and applied probability, emphasising structural representation and methods in stochastic and scientific computing, with applications including model-based inference and uncertainty quantification. Recent topics include:

Markov chains and their representations (diffusions, particle systems, random fields, and walks)
Monte Carlo methods combined with computational algebra, hypergeometric functions, and emerging computational paradigms
Exchangeability in Bayesian-nonparametric and extreme-value modelling
Partial differential equations and differential geometry in estimation

Papers

A class of gaussian fields on Z^d_q, with Robert Griffiths. arXiv: 2602.19469
Gaussian free fields on Hamming graphs and lattice spin systems. arXiv: 2512.24199
Random walks on Z^d_q, with Robert Griffiths. arXiv: 2510.22554
Spectral analysis of hierarchical continuous-time quantum walks, with Jiro Akahori, Yusuke Ide, Tomoki Kato, Norio Konno, and Akihiro Narimatsu. arXiv: 2510.12043
Symmetric quantum walks on Hamming graphs and their limit distributions, with Robert Griffiths. arXiv: 2509.26243
Algorithm for direct sampling from conditional distributions of toric models, with Nobuki Takayama. arXiv: 2110.14922
Asymptotic bias reduction of maximum likelihood estimates via penalized likelihoods with differential geometry, with Masayo Y. Hirose. Bernoulli, to appear. arXiv: 2011.14747
Direct sampling from conditional distributions by sequential maximum likelihood estimations. Algebraic Stat. 16(2): 175-199 (2025) arXiv: 2502.00812 paper poster
Representations of finite Markov chains and direct sampling. J. Japan Stat. Soc. Ser. J. 55(1): 177-194 (2025) paper (in Japanese with English abstract)
Asymptotic UMVUE: Asymptotic moments matching the UMVUE under the Ewens sampling formula, with Masayo Y. Hirose. Calcutta Stat. Assoc. Bull. 75(2): 197-219 (2023) paper
A measure-on-graph-valued diffusion: a particle system with collisions and its applications. Mathematics. Special Issue: Random Combinatorial Structures,10: 4081 (2022)
Parameter estimation of generalized beta distributions and its application to a historical tsunami magnitude dataset, with Masaaki Sibuya. Pioneering Works on Extreme Value Theory: In Honor of Professor Masaaki Sibuya, 1-25, Springer, 2021

Book

Partitions, Hypergeometric Systems, and Dirichlet Processes in Statistics, Springer, 2018. errata

Miscellanea

Genji-ko is a clusterling: around Bell's number. Movie at Kyoto University (in Japanese). A public lecture in 2021 around integer partitions: from a classic art of refinement to deep learning
Recognition, a recreation in Bayesian updating. Column of Inst. Stat. Math. News 129 (2015)

Announcement

To those who live abroad and wish to study as a student of SOKENDAI

If you are applying for the MEXT scholarship with a university recommendation, the deadline is the beginning of July of the year prior to your intended enrollment year. If you hope to apply for the October 2027 admission, the admission period will be from the end of June to the beginning of July 2026. If you would like me to serve as your supervisor, please contact me by the specified deadline. If you are applying with an embassy recommendation, you may proceed without contacting me; however, you will be asked to contact me if you pass the initial screening.

Last updated: March 9, 2026