Introduction

This is a home page of a Japanese statistician Kei Kobayashi.

[Latest Affiliation]

Professor of Department of Mathematics, Faculty of Science and Technology, Keio University

Until March 2016, Kei Kobayashi belonged to The Research Organization of Information and Systems (ROIS) The Institute of Statistical Mathematics (ISM)

News

2024/10/16
Okudo, K. and Kobayashi, K.: Benign Overfitting under Learning Rate Conditions for α Sub-exponential Input, arXiv:2409.00733
2023/08/23
A session talk at 10th International Congress on Industrial and Applied Mathematics (ICIAM2023).
2023/06/30
Organizing commitee of Boston-Keio-Tsinghua Wokshop 2023 with a talk．（2023/06/26-2023/6/30）
2023/05/20
Invited speech at 80th Conference of the Japan Economic Policy Association.
2023/04/01
I become a professor of Keio University.
2022/05/23
Visiting Fellow of Data Science Institute in London School of Economics（2022/5/23-2022/9/16）
2022/05/16
Keynote speech at Algebraic Statistics 2022
2021/10/13
Co-editor of Annals of Institute of Statistical Matheamtics(AISM). (2021.10.1-)
2021/02/26
Tong, Q. and Kobayashi, K. (2021): Entropy-regularized optimal transport on multivariate normal and q-normal distributions, Entropy, 23(3), 302. link
2021/02/16
Takehara, D. and Kobayashi, K.: Enhancing Hierarchical Information by Using Metric Cones for Graph Embedding, arXiv:2102.08014
2020/02/29
Kamoi, R. and Kobayashi, K.: Why is the Mahalanobis Distance Effective for Anomaly Detection?, arXiv2003.00402
2020/01/30
Kobayashi, K. and Wynn, H. (2020), Empirical geodesic graphs and CAT(k) metrics for data analysis, Statistics and Computing, 30(1), 1-18, DOI: 10.1007/s11222-019-09855-3 link
2019/11/15
Kamoi, R. and Kobayashi, K.: Likelihood Assignment for Out-of-Distribution Inputs in Deep Generative Models is Sensitive to Prior Distribution Choice, arXiv1911.06515
2019/07/23
[Conference] Kobayashi, K.: Statistical inference and data analysis on length metric spaces, 32nd European Meeting of Statisticians, Palermo, 2019.7.23.
2019/06/18
Associate Editor of Annals of Institute of Statistical Matheamtics(AISM) (2019.6.18-2021.9.30)
2019/01/01
Chair of Grade Pre-1 of Japan Statistical Society Certificate (2019.1.1-2019.12.31)
2018/07/13
[Conference] Kobayashi, K. and Wynn, H.: Element-wise shrinkage of positive correlation matrices for increasing sparsity, 2018 IMS Annual Meeting on Probability and Statistics, Vilnius, Lithuania, 2018.07.03.(poster)
2018/03/18
Visiting researcher of Isaac Newton Institute Program "Uncertainty quantification for complex syms: theory and methodologies"(-2018/03/30)
2018/02/12
Aoshima, T., Kobayashi, K. and Minami, M.: Revisiting the Vector Space Model: Sparse Weighted Nearest-Neighbor Method for Extreme Multi-Label Classification, arXiv1802.03938
2017/09/05
[Conference] Kobayashi, K. and Wynn, H.: A new aspect of geometrical data analysis using curvature of the data space and the empirical graph, Royal Statistical Society Conference 2017, Glasgow, 2017.9.5.(poster)
2017/06/01
I become a visiting researcher of RIKEN AIP project.
2016/07/22
Shinzato, T., Kobayashi, K., Kaku, I. (2016) Universality of Makespan in Flowshop Scheduling Problem, arXiv1607.07303
2016/06/28
[Conference] Kei Kobayashi, Data analysis using curvature of data spaces and their metric cones, The 4th Institute of Mathematical Statistics Asia Pacific Rim Meeting(IMS-APRM2016), The Chinese University of Hong Kong.
2016/06/23
[Conference] Kei Kobayashi, Generating statistically efficient estimators via computational algebra, Application of Algebraic Methods to Statistics, RIMS, Kyoto University (invited).
2016/04/01
Kei Kobayashi becomes an associate professor of Department of Mathematics, Faculty of Science and Technology, Keio University.
2015/06/09
[Conference] Kobayashi, K. and Wynn, H., Intrinsic and extrinsic means and curvature of metric cones, Algebraic Statistics 2015, Genoa
2015/04/02
[Conference] Kobayashi, K., Geodesic distances on data spaces: their computation and modification, ISI-ISM-ISSAS joint Conference 2015, Tokyo
2014/10/01
I become research director of a JST PRESTO project "Statistial Analysis and Theory via Geometric Features of Data Spaces" in "Collaborative Mathematics for Real World Issues".
2014/09/22
[Conference] Kei Kobayashi, Mitsuru Orita and Henry P. Wynn, Statistical analysis via the curvature of data space, MaxEnt 2014, Amboise, France.
2014/07/09
Kobayashi, K. and Wynn, H. (2014) Computational algebraic methods in efficient estimation, Geometric Theory of Information (Signals and Communication Technology), Frank Nielsen ed., Springe-Verlag, 119-140. arXiv1310.6515
2014/07/09
[Conference] Kei Kobayashi and Henry Wynn, The empirical geodesic graphs and their deformation for data analysis, ASC-IMS 2014, Australian Technology Park, Sydney.
2014/03/14
[Conference] Kei Kobayashi, Curvature of empirical metrics on a data space and its deformation (work with Henry P. Wynn), Workshop on Mathematical Approaches to Large-Dimensional Data Analysis, ISM, Tokyo.
2014/03/12
Kei Kobayashi, Mitsuru Orita (2014), Permutation test for dendrograms and its application to the analysis of mental lexicons, arXiv1403.2845.
2014/02/20
[Conference] Kei Kobayashi, Hypothesis Testing for the Difference of Dendrograms (work with Mitsuru Orita), ISI-ISM-ISSAS Joint Conference 2014, ISI, Delhi.
2014/01/23
[Conference] Kei Kobayashi, Henry P. Wynn, The empirical geodesic graphs and a deformation of their metric Computational Algebraic Statistics, Theories and Applications (CASTA) 2014, Kyoto.
2014/01/13
Kei Kobayashi, Henry P. Wynn (2014), Empirical geodesic graphs and CAT(k) metrics for data analysis, arXiv1401.3020

E-mail

(My First Name)@math.keio.ac.jp