Caleb Ju
- Program Year: 4
- Academic Institution: Georgia Institute of Technology
- Field of Study: Operations Research
- Academic Advisor: Guanghui (George) Lan
- Practicum(s):
National Renewable Energy Laboratory (2022)
Lawrence Berkeley National Laboratory (2024) - Degree(s):
B.S. Mathematics and Computer Science, University of Illinois at Urbana-Champaign, 2020 - Personal URL: https://jucaleb4.github.io/
Summary of Research
I am interested in developing fast and scalable optimization algorithms for challenging problems arising in machine learning and computational science.Publications
- Caleb Ju and Guanghui Lan. Strongly-Polynomial Time and Validation Analysis of Policy Gradient.Methods. Pre-print, Sep 2024
- Caleb Ju and Guanghui Lan. Policy Optimization over General State and Action Spaces. Pre-print, Sep 2024.
- Caleb Ju and Constance Crozier. Learning a Local Trading Strategy: Deep Reinforcement Learning for Grid-scale Renewable Energy Integration. Accepted to Hawaii International Conference on System Sciences (HICSS-58), Aug 2024.
- Ji Gao, Abigael Whalen, Caleb Ju, Yongsheng Chen, Guanghui Lan, Zhaohui Tong. Reinforcement Learning-Based Control for Waste Biorefining Processes Under Uncertainty. Submitted, Jul 2023
- Caleb Ju and Guanghui Lan. Dual dynamic programming for stochastic programs over an infinite horizon. arXiv, Mar 2023
- Caleb Ju, Georgios Kotsalis, Guanghui Lan. A model-free first-order method for linear quadratic regulator with tilde{O}(varepsilon^{-1}) sampling complexity. Submitted, Dec 2022
- Caleb Ju, Serif Yesil, Mengyuan Sun, Chandra Chekuri, Edgar Solomonik. Efficient parallel implementation of the multiplicative weight update method for graph-based linear programs. arXiv, Aug 2021
- Yan Li, Caleb Ju, Ethan X. Fang, Tuo Zhao. Implicit regularization of Bregman proximal point algorithm and mirror descent on separable data. arXiv, Aug 2021
- Caleb Ju, Yifan Zhang, and Edgar Solomonik. Communication lower bounds for nested bilinear algorithms. Foundations of Computational Mathematics. Nov 2023 (https://link.springer.com/article/10.1007/s10208-023-09633-8)
- Caleb Ju and Edgar Solomonik. "Derivation and analysis of fast bilinear algorithms for convolution". SIAM Review. Nov 2020 (https://epubs.siam.org/doi/abs/10.1137/19M1301059)
Awards
- SIAM LA24 Travel Award- SIAM PP24 Travel Award
- 2023 INFORMS Annual Meeting Best Poster Finalist
- MOPTA 2023 Best Poster Award
- OP21 Travel Award
- Illinois Summer Research Poster: Best Research Presentation Award (Illinois)
- Franz Hohn and J.P. Nash Scholarship (Illinois)
- Dean's List, Fall 2017 (Illinois)