I am a postdoctoral researcher at the Center for Advanced Intelligence Project (AIP), RIKEN, Japan in the Continuous Optimization Team led by Prof. Akiko Takeda. Before my current position, I worked with Prof. Ching-pei Lee as a postdoctoral scholar at Academia Sinica. Prior to this, I held a postdoctoral fellowship at the Department of Mathematics, National Taiwan Normal University, where I also completed my Ph.D. in Mathematics under the guidance of Prof. Jein-Shan Chen. For a more detailed information about my background, please refer to my CV.
I am interested in the analysis and development of solution methods for nonconvex problems that arise from
My recent works are concerned with problems inspired by machine learning applications, including best subset selection, sparse regression, compressive sensing, and hyperparameter learning. In general, I work on developing practical algorithms for nonconvex optimization to better handle large-scale problems, which is done either through the formulation of new algorithms or by novel acceleration techniques for existing ones.
I am also interested in studying the global convergence of classical projection algorithms for feasibility problems, which are not fully understood for nonconvex settings. I am likewise involved with designing new algorithms for variational inequalities and complementarity problems, such as dynamic systems approaches, smoothing techniques, and more recently, projection methods for their feasibility reformulations.
Global convergence and acceleration of projection methods for feasibility problems involving union convex sets
Jan Harold Alcantara and Ching-pei Lee
accepted by Journal of Optimization Theory and Applications (2024)
[arXiv]
Unified smoothing approach for best hyperparameter selection problem using a bilevel optimization strategy
Jan Harold Alcantara, Chieu Thanh Nguyen, Takayuki Okuno, Akiko Takeda and Jein-Shan Chen
Mathematical Programming (2024)
[Link] [arXiv]
Method of alternating projections for the general absolute value equation
Jan Harold Alcantara, Jein-Shan Chen and Matthew K. Tam
Journal of Fixed Point Theory and Applications, vol. 25, Article 39 (2023)
[Link] [arXiv]
Accelerated projected gradient algorithms for sparsity constrained optimization problems
Jan Harold Alcantara and Ching-pei Lee
Advances in Neural Information and Processing Systems (NeurIPS 2022), vol. 35, pp. 26723–26735 (2022)
[Link] [arXiv]
A new class of neural networks for NCPs using smooth perturbations of the natural residual function
Jan Harold Alcantara and Jein-Shan Chen
Journal of Computational and Applied Mathematics, vol. 407, 114092 (2022)
[Link]
A neural network based on the metric projector for solving SOCCVI problem
Juhe Sun, Weichen Fu, Jan Harold Alcantara and Jein-Shan Chen
IEEE Transactions on Neural Networks and Learning Systems, vol. 32, no. 7, pp. 2886-2900 (2021)
[Link]
Smoothing strategy along with conjugate gradient algorithm for signal reconstruction
Caiying Wu, Jing Wang, Jan Harold Alcantara and Jein-Shan Chen
Journal of Scientific Computing, vol. 87, no. 1, Article 21 (2021)
[Link]
A novel generalization of the natural residual function and a neural network approach for the NCP
Jan Harold Alcantara and Jein-Shan Chen
Neurocomputing, vol. 413, pp. 368-382 (2020)
[Link]
On construction of new NCP functions
Jan Harold Alcantara, Chen-Han Lee, Chieu Thanh Nguyen, Yu-Lin Chang and Jein-Shan Chen
Operations Research Letters, vol. 48, pp. 115-121 (2020)
[Link]
Neural networks based on three classes of NCP-functions for solving nonlinear complementarity problems
Jan Harold Alcantara and Jein-Shan Chen
Neurocomputing, vol. 359, pp. 102-113 (2019)
[Link]
A four-operator splitting algorithm for
nonconvex and nonsmooth optimization
Jan Harold Alcantara, Ching-pei Lee and Akiko Takeda
arXiv:2406.16025 (2024)
Theoretical smoothing frameworks for general nonsmooth bilevel problems
Jan Harold Alcantara and Akiko Takeda
arXiv:2401.17852 (2024)