Scholar
Tomoyuki Obuchi
Google Scholar ID: AYqKJtcAAAAJ
Associate professor, Kyoto University
Statistical physics
machine learning
high-dimensional inference
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
362
H-index
10
i10-index
11
Publications
20
Co-authors
25
list available
Publications
8 items
Understanding Head Geometry and Dynamics in Federated Regression through a Natural Solution Selection Rule: An Unconstrained Feature Model Analysis
2026
Cited
0
Mechanistic Diagnostics of Spatial Lexical Bias in Multimodal Large Language Model Spatial Reasoning
2026
Cited
0
Perfect reconstruction of sparse signals using nonconvexity control and one-step RSB message passing
2025
Cited
0
Graphical model for tensor factorization by sparse sampling
2025
Cited
0
Neural Collapse in Cumulative Link Models for Ordinal Regression: An Analysis with Unconstrained Feature Model
2025
Cited
0
Analysis of High-dimensional Gaussian Labeled-unlabeled Mixture Model via Message-passing Algorithm
arXiv.org · 2024
Cited
0
Transfer Learning in ℓ1 Regularized Regression: Hyperparameter Selection Strategy based on Sharp Asymptotic Analysis
arXiv.org · 2024
Cited
0
When resampling/reweighting improves feature learning in imbalanced classification?: A toy-model study
arXiv.org · 2024
Cited
0
Resume
Research Experience
Dec 2019 – Present: Associate Professor, Department of Systems Science, Graduate School of Informatics, Kyoto University
Apr 2016 – Nov 2019: Assistant Professor, School of Computing, Tokyo Institute of Technology
Oct 2013 – Mar 2016: Assistant Professor, Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology
Sep 2012 – Sep 2013: Postdoctoral researcher, Department of Theoretical Physics, École Normale Supérieure Paris
Apr 2012 – Sep 2013: Cybermedia Center, Osaka University
Apr 2010 – Mar 2012: Graduate School of Science, Department of Earth and Space Science, Osaka University
Apr 2008 – Mar 2010: Graduate School of Science and Engineering, Tokyo Institute of Technology
Co-authors
8 total
Yoshiyuki Kabashima
Institute for Physics of Intelligence & Department of Physics, The University of Tokyo
MASATO OKADA
東京大学
Takuya Isomura
RIKEN Center for Brain Science
Yu Terada
Sony
Toshiyuki Tanaka
Graduate School of Informatics, Kyoto University
Remi Monasson
CNRS & Ecole Normale Supérieure
Xiangming Meng
The Zhejiang University-University of Illinois Urbana-Champaign Institute, Zhejiang University
Ulisse Ferrari
CRNS tenured, Institut de la Vision