Dongruo Zhou

Dongruo Zhou is an Assistant Professor in the Department of Computer Science at Indiana University Bloomington. His research focuses on foundational aspects of machine learning, particularly sequential decision-making (bandits, reinforcement learning), optimization algorithms for deep learning, and statistical complexity analysis. He holds a Ph.D. from UCLA (2023), an M.Sc. from Tsinghua University (2017), and previously studied at the University of Virginia (2017-2018).

Research interests span reinforcement learning theory (sample complexity, horizon-free algorithms), optimization in deep learning (stochastic gradient methods, saddle-point escape), and neural contextual bandits. Notable contributions include establishing nearly optimal bounds for RL with function approximation and analyzing adversarial corruption robustness in contextual bandits.

Recent work emphasizes decision-making algorithms for complex structures like large language models and hierarchical RL. His publications appear in top venues like ICML, NeurIPS, and COLT, with over 50 peer-reviewed papers. He advises three Ph.D. students at IU Bloomington.

His research has received recognition through prestigious awards including the NeurIPS 2022 Spotlight presentation and NeurIPS 2021 oral presentation. Key contributions include foundational work on reward-free exploration frameworks and variance-dependent regret bounds.