Teaching
- Introduction to Computer Programming (CSCI-SHU 11), NYU Shanghai, Fall 2024 — Laboratory Leader
- Advanced Topics in AI and Machine Learning (CS-UH 3212), NYU Abu Dhabi, Spring 2024 & Spring 2025 — Guest Lecturer on Imitation Learning and Offline Reinforcement Learning
Correspondence: zixuandong@nyu.edu
Zixuan is a Ph.D. student in Computer Science at NYU Shanghai and NYU Courant, advised by Professor Keith Ross. His research interests span the theory and application of reinforcement learning (RL) and deep learning, with a primary focus on improving the sample efficiency and generalization of deep RL algorithms for robotic control. Prior to his Ph.D., he earned a B.Sc. in Honors Mathematics and Data Science with a concentration in AI from NYU Shanghai. Outside of research, h enjoys practicing Kendo and cooking.
Publications and Preprints
- George Andriopoulos*, Zixuan Dong*, Bimarsha Adhikari*, Keith Ross, “Geometric Analysis of Neural Regression Collapse via Intrinsic Dimension”, arXiv preprint, 2026
- Zixuan Dong*, Yumi Omori*, Keith Ross, “Minimal Ingredients for Reward Assignment from Expert Demonstrations”, Accepted by RLC 2026
- George Andriopoulos*, Zixuan Dong*, Bimarsha Adhikari*, Keith Ross, “Geometric Properties of Neural Multivariate Regression: An Empirical Study”, Accepted by ICLR 2026 Sci4DL Workshop
- Yumi Omori*, Zixuan Dong*, Keith Ross, “Should We Ever Prefer Decision Transformer for Offline Reinforcement Learning?”, Accepted by RLC 2025 RLBrew Workshop
- Li Guo, George Andriopoulos, Zifan Zhao, Zixuan Dong, Shuyang Ling, Keith Ross, “Cross Entropy versus Label Smoothing: A Neural Collapse Perspective”, Accepted by TMLR, 2025
- George Andriopoulos*, Zixuan Dong*, Li Guo*, Zifan Zhao*, Keith Ross*, “The Prevalence of Neural Collapse in Neural Multivariate Regression”, Accepted by NeurIPS 2024
- Zecheng Wang*, Che Wang*, Zixuan Dong*, Keith Ross, “Pre-training with Synthetic Data Helps Offline Reinforcement Learning”, Accepted by ICLR 2024
- Zixuan Dong, Che Wang, Keith W. Ross, “On the Convergence of Monte Carlo UCB for Random-Length Episodic MDPs”, Undergraduate Thesis, 2022
(* Equal Contribution)
Service
- Reviewer for NeurIPS 2025, ICLR 2026, ICRA 2026, ICML 2026 (Gold Reviewer Award), NeurIPS 2026
Research Interests
- Deep Reinforcement Learning
- Robotics