Naoki Yoshida

Papers

Publication

• N.Yoshida, S.Nakakita, M.Imaizumi, "Effect of Random Learning Rate: Theoretical Analysis of SGD Dynamics in Non-Convex Optimization via Stationary Distribution", Transactions on Machine Learning Research, 2025. paper movie code

• N.Yoshida, S.Watanabe, "Upper Bound of Real Log Canonical Threshold of Tensor Decomposition and its Application to Bayesian Inference", International Symposium on Stochastic Systems Theory and Its Applications, 2023. paper

Workshop

• N.Yoshida, S.Nakakita, M.Imaizumi, "Effect of Random Learning Rate: Theoretical Analysis of SGD Dynamics in Non-Convex Optimization via Stationary Distribution", ICML Workshop on High-dimensional Learning Dynamics (HiLD), Vienna (Austria), 2024/07. paper

Education

• PhD: The University of Tokyo, Graduate School of Arts and Sciences, 2024/04-present

Supervisor: Masaaki Imaizumi

• Master (Science): Tokyo Institute of Technology, School of Computing, 2022/04-2024/03

Supervisor: Sumio Watanabe

• Bachelor (Engineering): The University of Tokyo, Faculty of Engineering, 2017/04-2022/03

Awards & Funding

• JSPS Doctoral Course Research Fellowships (DC2), 2025/04-present

• JST SPRING GX, 2024/04-2025/03

Other Experience

• Research internship in Sony Group Corporation, 2025/05-2025/10

• Machine learning engineer (internship) in Aces, 2022/04-2024/03

• Data scientist (internship) in Matsuo Institute, 2022/04-2024/03