Sho Sakai
Ph.D. Student in Mathematics
University of Tsukuba
My research focuses on high-dimensional statistical analysis, with interests extending beyond theoretical development to applications involving real-world data and decision-making processes.
I host a podcast titled “Data Science LG: Learning Together in Statistics and Data Science”, where we explore topics such as statistics, machine learning, and academic careers from the perspective of students and researchers. The podcast is available on Apple Podcast, Spotify, YouTube, and Amazon Music. For more details, please see the Community / Podcast section below.
I am also involved in advocacy to raise awareness about hematopoietic stem cell transplantation, sharing my experience as a donor and promoting donor registration through talks and outreach activities. Through these efforts, I aim to help bridge the gap between healthcare and society.
I share my research findings and code on GitHub, and post about blood donation, hematopoietic stem cell transplantation, data science, education, and research on note.
Interests
Table of Contents
Skill
Speech
Sho Sakai, Kazuyoshi Yata, and Makoto Aoshima, “Exploring Principal Component Regression in High-Dimensional Data: Hypothesis Testing for PCR Coefficients & Steps Toward Prediction-Error Minimisation”, ISACT 2026, Tokyo, Japan, Poster Presentation, February 16-17, 2026.
View ProgramSho Sakai, Kazuyoshi Yata, and Makoto Aoshima, “Exploring Principal Component Regression in High-Dimensional Data”, 異分野異業種研究交流会2025, Tokyo, Japan, Poster Presentation, October 25, 2025.
View ProgramSho Sakai, Kazuyoshi Yata, and Makoto Aoshima, “Exploring Principal Component Regression in High-Dimensional Data: Hypothesis Testing for PCR Coefficients & Steps Toward Prediction-Error Minimisation”, Statistics Summer Seminar 2025, Kagawa, Japan, Poster Presentation, August 4-6, 2025.
View ProgramSho Sakai, Kazuyoshi Yata, and Makoto Aoshima, “Hypothesis testing for PCR coefficients in high-dimensional data”, The Mathematical Society of Japan Annual Meeting, Tokyo, Japan, Oral Presentation, March 21, 2025.
View ProgramSho Sakai, Kazuyoshi Yata, and Makoto Aoshima, “Hypothesis testing for PCR coefficients in high-dimensional data”, Seminars by Alumnae/Alumni of Kagoshima University on their Recent Achievements 2025, Kagoshima, Japan, Oral Presentation, March 11, 2025.
View ProgramEducation
University of Tsukuba
Ph.D. in Mathematics, Graduate School of Science, Degree Programs in Pure and Applied Sciences
Advisor: Aoshima Laboratory
University of Tsukuba
M.Sc. in Mathematics, Graduate School of Science, Degree Programs in Pure and Applied Sciences
Advisor: Aoshima Laboratory
Kagoshima University
B.Sc. in Mathematics and Informatics, Faculty of Science, Mathematics and Informatics Program
Advisor: Yoshida Laboratory (2022–2023)
Experience
Data Scientist (Contract work) — Nospare Inc.
Worked as an independent contractor on educational programs and data science-related projects.
Next-Generation AI Human Resource Development Program Fellow — JST BOOST initiative, University of Tsukuba
Selected for the University of Tsukuba's Next-Generation AI Talent Development Program, "Project for Interdisciplinary Next-Generation AI Innovative Human Resource Development". This initiative, part of a national strategy to foster doctoral students, supports 600 students across Japan. The University of Tsukuba has a quota of 12 students, and I was chosen as one of them through a highly competitive internal selection process. Receiving support for living expenses and research funds, I am able to fully dedicate myself to theoretical and applied research in high-dimensional statistics. I plan to actively pursue collaborations with fields such as bioinformatics, materials science, and astronomy.
Chair & Technical Assistant — User Committee, Department of Mathematics, University of Tsukuba
Appointed chair and technical assistant of the graduate student user committee for computing services, supporting computational infrastructure and student system administration.
International Symposium Staff — International Symposiums on Large Complex Data
Contributed to the 2023 International Symposium on Recent Advances in Theories and Methodologies for Large Complex Data and the 2024 International Symposium on Theories, Methodologies and Applications for Large Complex Data. Reported each activity on LinkedIn.
Staff, Mathematics Trial Program — University of Tsukuba
Served as staff for the undergraduate mathematics experience program in July 2023 and July 2024.
Intern
Research Assistant
University laboratory
Oct. 2023 - Nov. 2024
Award
Bone Marrow Donation Certificate of Appreciation（Minister of Health, Labour and Welfare）
Community / Podcast
2024 - present
I lead the design and management of study group templates, including topics such as multivariate analysis, reinforcement learning, Gaussian processes, and Bayesian deep learning. I also participate in the practical machine learning study group.
I host a podcast titled 'Data Science LG: Learning Together in Statistics and Data Science', where we explore topics such as statistics, machine learning, and academic careers from the perspective of students and researchers. The podcast is available on Spotify, Apple Podcast, YouTube, and Amazon Music.
For an overview of our podcast activities and related information, please visit the Notion page below:
Cast / Guest List
Titles and affiliations are as of the time of recording. Apple Podcast links are provided for each entry; the show is also available on Spotify, YouTube, and Amazon Music. * indicates guests who have appeared in multiple episodes. The episode link is to their first appearance; you can find other episodes by searching for their name on each platform.
| 名前 | 所属 | SNSなど | エピソード |
|---|---|---|---|
| 津志田 侑弥さん* | 立正大学データサイエンス学部 2021年度入学 | 🔗 | |
| 中山 優吾 さん* | 企業研究員 | HPLinkedIn | 🔗 |
| 岩永 悠希さん | 神戸大学大学院経済学研究科 修士２年 | HP | 🔗 |
| 上野 孝斗 さん* | 滋賀大学大学院データサイエンス研究科 修士2年 | X | 🔗 |
| 久保 知生さん | dip株式会社 | XLink | 🔗 |
| 司馬 博文さん* | 総合研究大学院大学 先端学術院 統計科学コース 五年一貫博士課程 2年 →総合研究大学院大学統計科学コース5年一貫D3（2025年度より） | 発表資料XHP | 🔗 |
| 平木 大智さん* | 東京大学経済学研究科統計コースD1（2025年度より） | XLinkedIn発表内容の論文 | 🔗 |
| 薗部 成輝 さん | 東京理科大学 創域理工学研究科 情報計算科学専攻修士２年 | X公開されている論文論文について | 🔗 |
| 加藤 真大さん* | みずほ第一フィナンシャルテクノロジー データアナリティクス技術開発部 東京大学 博士後期課程 | XLinkedInWebサイト | 🔗 |
| 長谷川 弘貴さん* | 筑波大学大学院理工情報生命学術院システム情報工学研究群サービス工学学位プログラム 修士2年 | LinkedInHP | 🔗 |
| 井下 敬翔さん | 関西大学 大学院 商学研究科 商学専攻 博士後期課程 | HPResearchmapXLinkedIn | 🔗 |
| 戸簾 隼人さん* | 滋賀大学 大学院 データサイエンス研究科 博士後期課程 | HPLinkedInResearchmap | 🔗 |
| MC: 酒井 彰* | 筑波大学 大学院 博士後期課程 全ての回に出演しています。編集やゲストへの連絡などの運営を担当しています。 | Nospare Student Community 運営XLinkedInHP | — |
| MC: 北野 優斗* | 早稲田大学 大学院 修士課程 | Nospare Student Community 運営X | 🔗 |
| MC: 段 暁然* | EPSRC CDT in Statistics and Machine Learning at Imperial College London and the University of Oxford (博士課程) | X | 🔗 |
| 佐野 和幸さん | LINEヤフー株式会社 データサイエンティスト | XLinkedIn | 🔗 |
| 金髙 右京さん | Sansan株式会社 研究開発部 研究員 | XEightデジタル名刺 | 🔗 |
| 井口 亮さん | みずほ第一フィナンシャルテクノロジー データアナリティクス技術開発部長 | 🔗 | |
| 陳 鈺涵さん | みずほ第一フィナンシャルテクノロジー ファイナンシャルエンジニア | HP | 🔗 |
| 増田 伊吹 さん* | 筑波大学大学院 情報理工学位プログラム D1 | 🔗 |
Conference Reviews
Apple Podcast links are provided for each entry; the show is also available on Spotify, YouTube, and Amazon Music.
| 名前 | 日付 | エピソード | note |
|---|---|---|---|
| 計算技術による学際的統計解析ワークショップ（#14, 15） | 2025年6月11日 | 🔗 | — |
| 第19回日本統計学会春期集会（#17, 18） | 2025年6月11日 | 🔗 | — |
| 日本数学会2025年度年会（#19） | 2025年6月11日 | 🔗 | — |
| 日本応用数理学会2025年度年会（#41） | 2025年9月21日 | 🔗 | — |
| 白金鉱業Meetup vol.20 効果検証編（#42） | 2025年10月11日 | 🔗 | 🔗 |
| NeurIPS 2025、CMStat 2025、みずほ、統計サマーセミナー 2025、統計関連学会連合大会 2025、異分野異業種研究交流会 2025（#44） | 2025年12月28日 | 🔗 | — |
Podcast Reviews
Apple Podcast links are provided for each entry; the show is also available on Spotify, YouTube, and Amazon Music.
2024 - 2025
I was a general member of this graduate student community and started a podcast initiative within the group.
Outreach / Public Engagement
Achieved 70 blood donations and received the silver merit badge.
I engage in activities to share information about hematopoietic stem cell transplantation and promote donor registration. If you're interested in inviting a speaker or organizing a promotional booth about the Japan Marrow Donor Program at your school, university, or organization, please reach out. We can coordinate with the Japan Marrow Donor Program directly. To increase the number of young bone marrow donors, I believe it is important to share the voices of those who have actually donated and their families.
(Feb. 14, 2026) I appeared on Kamakura FM as a bone marrow donor | Passing the baton of life begins with awareness
(July 7, 2025) Gave a lecture as a bone marrow donor at a nursing school.
(May 28, 2025) A summary of bone marrow transplant / bone marrow bank / blood donation related podcasts
(May 11, 2025) Participated in Health Festival to promote awareness of hematopoietic stem cell transplantation
(Apr. 23, 2025) "Voices of Donor Families" featured
(Dec. 15, 2024) Gave a lecture as a bone marrow donor at the 'Egao Japan Marrow Donor Program Student Seminar'
(Dec. 2024) Interviewed by the Japan Marrow Donor Program on my donation experience and featured in their newsletter
Teaching Assistant
Qualification / Test Score
Japanese High School Teacher’s License (Information & Mathematics)
Mar. 2023
Japanese Junior High School Teacher’s License (Mathematics)
Mar. 2023
Mar. 2022
Academic Society
The Japanese Society for Artificial Intelligence
2025 – present
2023 – present
The Mathematical Society of Japan
Oct. 2023 – Mar. 2026
Portfolio
Master's Thesis / Hypothesis testing for PCR coefficients in high-dimensional data (JSS NEWS)
Introduced in JAPAN STATISTICAL SOCIETY NEWS (April 20, 2025, No.203, Section 9: p.15).
Master's Thesis / Hypothesis testing for PCR coefficients in high-dimensional data (Mathematics Communication)
Featured in Mathematical Society of Japan's 'Mathematics Communication' (Volume 30, No.1, May 2025) in the section '2024 Master's and Doctoral Theses'.
Graduation Thesis / Asymptotic Theory of Principal Component Analysis
Related repository link available above.
Asymptotic Theory of PCA
R code and mathematical notes on asymptotic properties of PCA under large-sample settings.
T.W. Anderson (2003) PCA Confirmatory Analysis
Reimplementation of confirmatory PCA methods from Anderson's 2003 paper.
T.W. Anderson (2003) Hypothesis Testing
Reproduction of hypothesis testing framework for PCA from Anderson (2003).
Principal Component Analysis
Notebooks to demonstrate PCA using both synthetic and real-world data. Implementations are available in Jupyter (Python) and R.
Conditional Probability and Multiplication Theorem
A TeX document that explains the relationship between conditional probability and multiplication rule.
PRML with Python
Selected implementations and exercises based on 'Pattern Recognition and Machine Learning' (Bishop).
Causal Inference
Jupyter notebooks exploring potential outcome frameworks and causal graphs.
Programming with R, C, and Python
Introductory examples in multiple languages for basic computational logic and syntax.
Transformation of Matrix
Utility scripts and explanations for matrix operations and transformations.