I study optimization algorithms for data-driven decision making under a dynamically changing environment. My most recent works have studied this through the lens of reinforcement learning and their application to energy and computational science.
Overview
Published
(α) Alphabetical order | *Equal contribution
-
Strongly-polynomial time and validation analysis of policy gradient methods
(α) Caleb Ju and Guanghui Lan
Mathematical Programming (in press)
📌 The Alice and John Jarvis Best Paper Award (2025)
[DOI] [preprint] [code] -
Policy optimization over general state and action spaces
(α) Caleb Ju and Guanghui Lan
SIAM Journal on Optimization
[DOI] [preprint] [code] -
A model-free first-order method for linear quadratic regulator with Õ(1/ε) sampling complexity
Caleb Ju, Georgios Kotsalis, Guanghui Lan
SIAM Journal on Control and Optimization
[DOI] [preprint] [code] -
Learning a local trading strategy: deep reinforcement learning for grid-scale renewable energy integration
Caleb Ju and Constance Crozier
Hawaii International Conference on System Sciences 2025
[DOI] [preprint] [code] [DEIXIS] -
Reinforcement learning-based control for waste biorefining processes under uncertainty
Ji Gao, Abigael Whalen, Caleb Ju, Yongsheng Chen, Guanghui Lan, Zhaohui Tong
Communications Engineering
[DOI] [preprint] -
Communication lower bounds for nested bilinear algorithms via rank expansion of Kronecker products
Caleb Ju*, Yifan Zhang*, Edgar Solomonik
Foundations of Computational Mathematics
[DOI] [preprint] -
Derivation and analysis of fast bilinear algorithms for convolution
Caleb Ju and Edgar Solomonik
SIAM Review
[DOI] [preprint] [code]
Working Papers
-
Auto-exploration for online reinforcement learning
(α) Caleb Ju and Guanghui Lan
Submitted
[preprint] [code] -
Reinforcement Learning for Distributed Control of Bi-directional Vehicle Charging
Claas-Christoph Heitzhausen, Caleb Ju, Constance Crozier
Submitted
-
Actor-accelerated policy dual averaging for reinforcement learning in continuous action spaces
Ji Gao, Caleb Ju, Guanghui Lan, Zhaohui Tong
Submitted
[preprint] -
Dual dynamic programming for stochastic programs over an infinite horizon
Caleb Ju and Guanghui Lan
Under revision
📌 MOPTA Best Poster Award (2023)
[preprint] [code] -
Preconditioning via Randomized Range Deflation (RandRAND)
Oleg Balabanov, Caleb Ju, Kaiwen He, Aryaman Jeendgar, Michael Mahoney
Preprint
[preprint] -
Efficient parallel implementation of the multiplicative weight update method for graph-based linear programs
Caleb Ju, Serif Yesil, Mengyuan Sun, Chandra Chekuri, Edgar Solomonik
Preprint
[preprint] -
Implicit regularization of Bregman proximal point algorithm and mirror descent on separable data
Yan Li, Caleb Ju, Ethan X. Fang, Tuo Zhao
Preprint
[preprint]