portrait

I am an Eric and Wendy Schmidt AI in Science Fellow at the Data Science Institute, University of Chicago.

My research interest is in analysis, design and application of Monte Carlo sampling methods.

I received my PhD in mathematics from the Courant Institute of Mathematical Sciences, New York University. I was fortunate to be advised by Prof. Jonathan Weare. Before my PhD study, I obtained my B.S. degree in Computational Mathematics at Peking University. I worked with Prof. Lei Zhang and Prof. Tiejun Li. I was mostly interested in mathematical modeling of biological systems, which in the end inspires me to study more on stochasticity.

Publications and Preprints

  • Xiaoou Cheng, Daniel Sanz-Alonso, Nathan Waniorek. Time-uniform accuracy of ensemble Kalman filters with localization. [arXiv]
    Preprint (2026).

  • Yifan Chen, Xiaoou Cheng, Jonathan Niles-Weed, Jonathan Weare. Delocalization of bias in unadjusted Hamiltonian Monte Carlo and underdamped Langevin. [arXiv]
    Preprint (2026).

  • Yifan Chen, Xiaoou Cheng, Jonathan Niles-Weed, Jonathan Weare. Convergence of Unadjusted Langevin in High Dimensions: Delocalization of Bias. [Publisher] [arXiv]
    Communications on Pure and Applied Mathematics (2026).

  • Xiaoou Cheng, Jonathan Weare. The surprising efficiency of temporal difference learning for rare event prediction. [Publisher] [OpenReview] [arXiv]
    Advances in Neural Information Processing Systems (NeurIPS 2024).

  • Atsushi Shimizu, Xiaoou Cheng, Christopher Musco, Jonathan Weare. Improved Active Learning via Dependent Leverage Score Sampling. [OpenReview] [arXiv]
    International Conference on Learning Representations (ICLR 2024).
    Invited for oral presentation.

  • Xiaoou Cheng, Maria R. D’Orsogna, Tom Chou. Mathematical Modeling of Depressive Disorders: Circadian Driving, Bistability and Dynamical Transitions. [Publisher]
    Computational and Structural Biotechnology Journal (2021).

Contact

  • Email: chengxo_at_uchicago_edu; chengxo_at_nyu_edu