Fan Zhou

Name: Zhou Fan

Title:  Professor

Research Directions: Deep Learning, Reinforcement Learning, Causal Inference, Large Models

E-mail： zhoufan@mail.shufe.edu.cn

Personal Profile:

Zhou Fan, associate professor at the School of Statistics and Data Science, Shanghai University of Finance and Economics, is a Young Chang Jiang Scholar of the Ministry of Education. He received his PhD from the University of North Carolina at Chapel Hill and currently serves as an associate editor for the top statistical journal JASA. His research interests include algorithms and theories of deep learning, reinforcement learning, spatiotemporal networks, and causal inference. He has published dozens of papers in top journals and conferences in statistics and machine learning, including the Journal of the American Statistical Association, Journal of Machine Learning Research, NeurIPS, ICML, and ICLR. He leads a team that established the world's first large model statistical reasoning dataset and evaluation system. He has received the New Researcher Award from the International Conference of the Chinese Statistical Association, the James E. Grizzle Distinguished Alumnus Award from the University of North Carolina at Chapel Hill, and the Barry H. Margolin Award.

Recruitment Target:

PhD Student Admission Requirements: Aspiring to work in academia in the future; solid mathematical foundation, strong self-motivation, and a passion for research; familiarity with Python and GPU computing, with strong programming skills preferred.

Master's Student Admissions Requirements: Candidates are expected to continue pursuing doctoral studies and research in the future, or to work in industry in positions related to algorithms and scientific research. Master's students in the group will participate in systematic research training led by doctoral students and will be recommended to top universities in China and abroad for further doctoral studies, or to intern in research departments related to AI.

Research Project

Research Achievements

* Corresponding author; † Supervised student

Part of journal articles:

[1] Bang Liu, Run Yang†, and Fan Zhou∗. Discussion of “LAMBDA: Large model based data agent”. Journal of the American Statistical Association, 2025.

[2] Chengchun Shi, Zhengling Qi∗, Jianing Wang†, and Fan Zhou∗. Value enhancement of reinforcement learning via efficient and robust trust region optimization. Journal of the American Statistical Association, 119(547):2011–2025, 2024.

[3] Xingdong Feng, Yuling Jiao, Lican Kang, Baqun Zhang, and Fan Zhou∗. Over-parameterized deep nonparametric regression for dependent data with its applications to reinforcement learning. Journal of Machine Learning Research, 24(383):1–40, 2023.

[4] Fan Zhou, Shikai Luo, Xiaohu Qie, Jieping Ye, and Hongtu Zhu∗. Graph-based equilibrium metrics for dynamic supply–demand systems with applications to ride-sourcing platforms. Journal of the American Statistical Association, 116(536):1688–1699, 2021.

[5] Chenjia Bai, Ting Xiao, Zhoufan Zhu†, Lingxiao Wang, Fan Zhou, Animesh Garg, Bin He, Peng Liu, and Zhaoran Wang. Monotonic quantile network for worst-case offline reinforcement learning. IEEE Transactions on Neural Networks and Learning Systems, 2022.

[6] Fan Zhou, Haibo Zhou, Tengfei Li, and Hongtu Zhu. Analysis of secondary phenotypes in multigroup association studies. Biometrics, 76(2):606–618, 2020.

[7] Bingxin Zhao, Tianyou Luo, Tengfei Li, Yun Li, Jingwen Zhang, Yue Shan, Xifeng Wang, Liuqing Yang, Fan Zhou, Ziliang Zhu, et al. Genome-wide association analysis of 19,629 individuals identifies variants influencing regional brain volumes and refines their genetic co-architecture with cognitive and mental health traits. Nature Genetics, 51(11):1637–1644, 2019.

Conference Paper:

[1] Qi† Kuang, Jiayi Wang, Fan Zhou∗, and Zhengling Qi∗. Breaking the order barrier: Off-policy evaluation for confounded pomdps. In 39th Conference on Neural Information Processing Systems (NeurIPS 2025).

[2] Shuguang Yu†, Wenqian Xu†, Xinyi Zhou†, Xuechun Wang†, Hongtu Zhu, and Fan Zhou∗. Enhancing prediction performance through influence measure. In 13th International Conference on Learning Representations (ICLR 2025).

[3] Shuguang Yu†, Shuxing Fang†, Ruixin Peng†, Zhenglin Qi, Fan Zhou∗, and Chengchun Shi. Two-way deconfounder for off-policy evaluation under unmeasured confounding. In 38th Conference on Neural Information Processing Systems (NeurIPS 2024).

[4] Run Yang†, Yuling Yang†, Fan Zhou∗, and Qiang Sun∗. Directional diffusion model for graph representation learning. In 37th Conference on Neural Information Processing Systems (NeurIPS 2023).

[5] Ting Li, Chengchun Shi, Jianing Wang†, Fan Zhou, and Hongtu Zhu∗. Optimal dynamic treatment allocation for efficient policy evaluation. In 37th Conference on Neural Information Processing Systems (NeurIPS 2023).

[6] Qi Kuang†, Zhoufan Zhu†, Liwen Zhang, and Fan Zhou∗. Variance control for distributional reinforcement learning. In 40th International Conference on Machine Learning. (ICML 2023).

[7] Yang Sui†, Yukun Huang†, Hongtu Zhu, and Fan Zhou∗. Adversarial learning of distributional reinforcement learning. In 40th International Conference on Machine Learning. (ICML 2023).

[8] Sizhe Yu†, Ziyi Liu, Shixiang Wan, Jia Zheng, Zang Li, and Fan Zhou∗. Mdp2 forest: A constrained continuous multi-dimensional policy optimization approach for short-video recommendation. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022), pages 2388–2398, 2022.

[9] Fan Zhou, Zhoufan Zhu†, Qi Kuang†, and Zhang Liwen. Non-decreasing quantile function network with efficient exploration for distributional reinforcement learning. In 40th International Joint Conference on Artificial Intelligence (IJCAI 2021).

[10] Fan Zhou, Chenfan Lu†, Xiaocheng Tang, Fan Zhang, Zhiwei Qin, Jieping Ye, and Hongtu Zhu. Multi-objective distributional reinforcement learning for large-scale order dispatching. In 2021 IEEE International Conference on Data Mining (ICDM 2021), pages 1541–1546. IEEE, 2021.

[11] Fan Zhou, Jianing Wang†, and Xingdong Feng. Non-crossing quantile regression for distributional reinforcement learning. In 34th Conference on Neural Information Processing Systems (NeurIPS 2020), 33:15909–15919, 2020.

[12] Fan Zhou, Tengfei Li, Haibo Zhou, Hongtu Zhu, and Ye Jieping. Graph-based semi-supervised learning with non-ignorable non-response. In 33th Conference on Neural Information Processing Systems (NeurIPS 2019), 32, 2019.

Awards, Honors

2024 Ministry of Education Young Changjiang Scholars

2023 Shanghai Oriental Talent Youth Program

2022 Shanghai Morning Light Scholar

2021 Alzheimer's Disease Classification Challenge Global Runner-up, PRCV 2021

2020 Barry H. Margolin Award, Department of Biostatistics, UNC-Chapel Hill

2019 New Researcher Award, International Chinese Statistical Association (ICSA) Conference

2019 Travel Award for junior faculties, NeurIPS 2019

2017 The 1st place of the Grand Challenge, ISBI 2017