Shummin NAKAYAMA's Homepage

Profile

Name: Shummin Nakayama
Ph.D: Tokyo University of Science

researchmap

Google Scholar

Affiliation

Info-Powered Energy System Research Center (iPERC),
The University of Electro-Communications
Position: Assistant professor

Research Areas

  • Continuous optimization theory
  • Quasi-Newton method
  • Nonlinear conjugate gradient method
  • Proximal gradient method
  • DC Programming

Curriculum Vitae

Employment History

  • April 2019 -- March 2021: Assistant Professor
    Department of Industrial and Systems Engineering, Faculty of science and engineering,
    Chuo University

  • April 2021 -- September 2022: Assistant Professor
    Department of Data Science for Business Innovation, Faculty of science and engineering,
    Chuo University

  • October 2022 -- present: Assistant Professor
    Info-Powered Energy System Research Center (iPERC),
    The University of Electro-Communications

Academic Background

  • April 2010 -- March 2014: Department of Mathematical Information Science, Tokyo University of Science
  • April 2014 -- March 2016: Master Course in Department of Mathematical Science for Information Science, Tokyo University of Science
  • April 2016 -- March 2019: Doctor Course in Department of Applied Mathematics, Tokyo University of Science

Research Achievements

Refereed Papers in Journals

  1. Shotaro Yagishita and Shummin Nakayama, ''An acceleration of proximal diagonal Newton method", JSIAM Letters, 16 (2024), pp. 5--8. [DOI]
  2. Shummin Nakayama, Yasushi Narushima and Hiroshi Yabe, ''Inexact proximal DC Newton-type method for nonconvex composite functions", Computational Optimization and Applications, 87 (2024), pp. 611--640. [DOI,Code]
  3. Yasushi Narushima, Shummin Nakayama, Masashi Takemura, and Hiroshi Yabe, ''Memoryless Quasi-Newton Methods Based on the Spectral-Scaling Broyden Family for Riemannian Optimization", Journal of Optimization Theory and Applications, 197 (2023), pp. 639--664. [DOI]
  4. Yasushi Narushima and Shummin Nakayama, ''A proximal quasi-Newton method based on memoryless modified symmetric rank-one formula", Journal of Industrial and Management Optimization, 19 (2023), pp. 4095--4111. [DOI]
  5. Shummin Nakayama, Yasushi Narushima, Hiroaki Nishio, Hiroshi Yabe, ''An active-set memoryless quasi-Newton method based on a spectral-scaling Broyden family for bound constrained optimization", Results in Control and Optimization, 3 (2021), 100012. [DOI]
  6. Shummin Nakayama and Jun-ya Gotoh, ''On the superiority of PGMs to PDCAs in nonsmooth nonconvex sparse regression", Optimization Letters, 15 (2021), pp.2831--2860. [DOI]
  7. Shummin Nakayama, Yasushi Narushima and Hiroshi Yabe, ''Inexact proximal memoryless quasi-Newton methods based on the Broyden family for minimizing composite functions", Computational Optimization and Applications, 79 (2021), pp.127--154. [DOI]
  8. 成島康史，中山舜民，矢部博，''無制約最適化問題に対するメモリーレス準ニュートン法について (Memoryless quasi-Newton methods for unconstrained optimization)"，応用数理，29(4) (2019), pp. 8--17. [DOI, PDF]
  9. Shummin Nakayama ''A hybrid method of three-term conjugate gradient method and memoryless quasi-Newton method for unconstrained optimization", SUT Journal of Mathematics, 54 (2018), pp.79--98. [PDF]
  10. Shummin Nakayama, Yasushi Narushima and Hiroshi Yabe, ''Memoryless quasi-Newton methods based on spectral-scaling Broyden family for unconstrained optimization", Journal of Industrial and Management Optimization, 15 (2019), pp. 1773--1793. [DOI]
  11. Shummin Nakayama, Yasushi Narushima and Hiroshi Yabe, ''A memoryless symmetric rank-one method with sufficient descent property for unconstrained optimization", Journal of the Operations Research Society of Japan, 61 (2018), pp.53--70. [DOI, PDF]

Proceedings

  1. Shummin Nakayama, Yasushi Narushima, ''Global convergence of a proximal memoryless symmetric rank one method for minimizing composite functions", In Proceedings of International Conference on Nonlinear Analysis and Convex Analysis - International Conference on Optimization Techniques and Applications (NACA-ICOTA2019), (Hakodate, Japan, 2019), Contents II, pp. 99--108, 2021. [PDF]
  2. Hiroaki Nishio, Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''A globally convergent active-set memoryless quasi-Newton method based on spectral-scaling Broyden family for bound constrained optimization", In Proceedings of International Conference on Nonlinear Analysis and Convex Analysis - International Conference on Optimization Techniques and Applications (NACA-ICOTA2019), (Hakodate, Japan, 2019), Contents II, pp. 147--160, 2021. [PDF]

Conference Activities & Talks

  1. *Yasushi Narushima, Shummin Nakayama, Hiroshi Yabe, ''Nonmonotone proximal structured quasi-Newton methods based on the Broyden family ", The Third Pacific Optimisation Conference (POC2023), Sunway City, Malaysia, December 10, 2023.
  2. *Shummin Nakayama, Shotaro Yagishita, ''Proximal diagonal Newton method for nonconvex composite optimization", The Third Pacific Optimisation Conference (POC2023), Sunway City, Malaysia, December 10, 2023.
  3. *Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''Proximal structured quasi-Newton method for nonlinear least squares with nonsmooth regularizer", 10th International Congress on Industrial and Applied Mathematics (ICIAM), August 21, 2023.
  4. *Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''An Inexact Proximal Difference-of-Convex Algorithm Based on Memoryless Quasi-Newton Methods", 2021 SIAM Conference on Optimization, July 23, 2021 (Virtual Conference).
  5. *Jun-ya Gotoh, Shummin Nakayama, ''Continuous Exact Penalty Approach To Grouped Variable Selection In Regression Methods", INFORMS Annual Meeting, November 14, 2020 (Virtual Conference).
  6. *Jun-ya Gotoh, Shummin Nakayama, ''Sparse Robust Regression With Continuous Exact K-sparse Penalties", INFORMS Annual Meeting, Catonsville, USA, October 23, 2019.
  7. *Hiroaki Nishio, Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''Global convergence of an active-set memoryless quasi-Newton method based on spectral-scaling Broyden family for bound constrained optimization", International Conference on Nonlinear Analysis and Convex Analysis - International Conference on Optimization Techniques and Applications, Hakodate, Japan, August 30, 2019.
  8. *Shummin Nakayama, Yasushi Narushima, ''Global convergence of a proximal memoryless symmetric rank one method for minimizing composite functions", International Conference on Nonlinear Analysis and Convex Analysis - International Conference on Optimization Techniques and Applications, Hakodate, Japan, August 29, 2019.
  9. *Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''Inexact proximal memoryless quasi-Newton methods based on Broyden family for minimizing composite functions", The Sixth International Conference on Continuous Optimization of the Mathematical Optimization Society, Berlin, Germany, August 7, 2019.
  10. *Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''Inexact proximal memoryless spectral-scaling MBFGS mehtod", 23rd International Symposium on Mathematical Programming, Bordeaux, France, July 5, 2018.
  11. *Shummin Nakayama, Yasushi Narushima, Hiroshi Yabe, ''Global Convergence of Memoryless Quasi-Newton Methods Based on Broyden Family for Unconstrained Optimization", 2017 SIAM Conference on Optimization, Vancouver, Canada, May 24, 2017.
  12. *Shummin NAKAYAMA, Yasushi NARUSHIMA, Hiroshi YABE, ''A memoryless sized symmetric rank-one method with sufficient descent property for unconstrained optimization", The Fifth International Conference on Continuous Optimization of the Mathematical Optimization Society, Tokyo, Japan, August 11, 2016.

Contact

E-mail: snakayama<at>uec.ac.jp (Please replace <at> by @)