International Symposium on Theories, Methodologies and

Applications for Large Complex Data

Supported by
Grant-in-Aid for Scientific Research (A) 20H00576
"Innovative developments of theories and methodologies for large complex data"
(Principal Investigator: Makoto Aoshima)

Grant-in-Aid for Challenging Research (Exploratory) 22K19769
"Developments of statistical compression technology for massive data having tensor structures"
(Principal Investigator: Makoto Aoshima)


Date: December 4-6, 2024
Venue: Conference Room 202, Tsukuba International Congress Center
2-20-3 Takezono, Tsukuba, Ibaraki 305-0032, Japan
Contents & Purposes: Please look at THIS PAGE for further details.
Access: Please see THIS PAGE.
Dinner: December 5, 7:00pm-
Tenryu (URL: https://gurunavi.com/en/gf02801/rst/?sc_cid=r-foreign_en_service&__ngt__=TT15e83c984007ac1e4ae4b5YfxMOCr9NPtePXoHbz-SIU)

Program Schedule (PDF)/ Flyer (PDF)

* denotes speakers

December 4

1:40-1:50pm Opening

Time (UTC+9) Title and Speaker (Affiliation)
Invited Session
1:50-2:30pm Statistical inference on high-dimensional covariance structures under the SSE models    [Abstract]
   Aki Ishii*,a, Yumu Iwanab, Kazuyoshi Yatac and Makoto Aoshimac
  a(Department of Information Sciences, Tokyo University of Science)
  b(Graduate School of Science and Technology, University of Tsukuba)
  c(Institute of Mathematics, University of Tsukuba)
2:40-3:20pm Statistical properties of matrix decomposition factor analysis    [Abstract]
  Yoshikazu Terada
  (Graduate School of Engineering Science, Osaka University)
3:30-4:10pm High-dimensional statistics in astrophysics and its perspective    [Abstract]
  Tsutomu T. Takeuchi
  (Division of Particle and Astrophysical Science, Nagoya University)
4:30-5:10pm High-dimensional inference on a cross data matrix-based method    [Abstract]
  Shao-Hsuan Wang
  (Graduate Institute of Statistics, National Central University)
5:20-6:00pm On estimation of a matrix mean under matrix loss    [Abstract]
  Yuan-Tsung Chang*,a, Nobuo Shinozakib and William E. Strawdermanc
  a(The Institute of Statistical Mathematics)
  b(Faculty of Science and Technology, Keio University)
  c(Department of Statistics, Rutgers University)

December 5

Time (UTC+9) Title and Speaker (Affiliation)
Young Researchers Session
9:00-11:00am 1. Automatic sparse estimation of high-dimensional cross-covariance matrix    [Abstract]
  Tetsuya Umino
  (Graduate School of Science and Technology, University of Tsukuba)
2. Augmented estimation of principal component subspace in high dimensions    [Abstract]
  Dongsun Yoon
  (Department of Statistics, Seoul National University)
3. James-Stein estimator of spiked leading eigenvector of high-dimensional covariance matrix    [Abstract]
  Giheon Seong
  (Department of Statistics, Seoul National University)
4. General measures of attribution disclosure risk for gauging privacy of synthetic data    [Abstract]
  Yongjae Kim
  (Department of Statistics, Seoul National University)
5. Regularized k-POD clustering for high-dimensional missing data
  Guan Xin
  (Graduate School of Engineering Science, Osaka University)
Invited Session
11:10-11:50am Non-sparse high-dimensional statistics: structured model, neural network, and universality
  Masaaki Imaizumi
  (Komaba Institute for Science, The University of Tokyo / RIKEN AIP)
11:50-1:40pm Lunch
Special Invited Session
1:40-2:30pm Difference between large statistical model and medium statistical model    [Abstract]
Speaker: Shurong Zheng (School of Mathematics and Statistics, Northeast Normal University)
Discussion Leader: Kento Egashira (Department of Information Sciences, Tokyo University of Science)
2:40-3:30pm Principal component analysis for zero-inflated compositional data    [Abstract]
Speaker: Sungkyu Jung (Institute for Data Innovation in Science, Seoul National University)
Discussion Leader: Kazuyoshi Yata (Institute of Mathematics, University of Tsukuba)
Keynote Session
3:50-4:50pm A generalized mean approach for distributed-PCA    [Abstract]
Speaker: Su-Yun Huang (Institute of Statistical Science, Academia Sinica)
Discussion Leader: Yuan-Tsung Chang (The Institute of Statistical Mathematics)
5:00-6:00pm Alignment and matching tests for high-dimensional tensor signals via tensor contraction    [Abstract]
Speaker: Jianfeng Yao (School of Data Science, Chinese University of Hong Kong (Shenzhen))
Discussion Leader: Yuta Koike (Graduate School of Mathematical Sciences, The University of Tokyo)
7:00-9:00pm Dinner

December 6

Time (UTC+9) Title and Speaker (Affiliation)
Invited Session
9:00-9:40am On dimension-free concentration of logistic regression
  Shogo Nakakita
  (Komaba Institute for Science, The University of Tokyo)
9:50-10:30am Subspace recovery in winsorized PCA    [Abstract]
  Sangil Han
  (Institute for Data Innovation in Science, Seoul National University)
10:45-11:25am High-dimensional bootstrap and asymptotic expansion    [Abstract]
  Yuta Koike
  (Graduate School of Mathematical Sciences, The University of Tokyo)
11:35am-12:15pm On a test for assessing vector correlation for latent factor models in high-dimensional settings    [Abstract]
  Takahiro Nishiyama*,a, Masashi Hyodob and Shoichi Naritac
  a(Department of Business Administration, Senshu University)
  b(Faculty of Economics, Kanagawa University)
  c(Graduate School of Economics, Kanagawa University)
12:20-1:05pm Asymptotic locations of bounded and unbounded eigenvalues of sample correlation matrices of certain factor models – application to a components retention rule
  Yohji Akama
  (Mathematical Institute, Tohoku University)

1:05-1:10pm Closing