About
I am an assistant professor in the Department of Statistics at the University of Washington, where I hold the Dorothy Gilford Early Career Endowed Professorship in Mathematical Statistics. Before joining UW in 2025, I was a postdoctoral researcher at ETH Zürich, working with Peter Bühlmann and Nicolai Meinshausen. I received my PhD in Mathematics from HKUST in 2022, advised by Tong Zhang, and my BS in Statistics from Fudan University in 2018.
My current research focuses on distributional learning, which leverages generative models to estimate complex probability distributions, bringing modern machine learning tools to core statistical problems and scientific applications. This agenda has three strands:
- Methods and theory. Engression and its extensions for scalable nonparametric distributional regression, with theoretical guarantees (engression, generalized engression models, reverse Markov learning, error analysis).
- Statistical problems. Distributional approaches to causal inference (instrumental variables, causal data simulation), robustness to distribution shifts, missing data, simulation-based inference, and dimension reduction (distributional autoencoder, sufficient dimension reduction).
- Scientific applications. Generative emulation and downscaling of climate models, uncertainty quantification in hydrology, and prediction of the effects of unseen gene perturbations and drug combinations.
Prospective students. I am happy to work with students. If you are interested in working with me, please apply directly to the PhD program in Statistics at UW. I apologize if I cannot respond to emails; this does not reflect a lack of interest.
Publications & preprints
Selected
- Engression: Extrapolation through the Lens of Distributional Regression X. Shen, N. Meinshausen Journal of the Royal Statistical Society Series B, 2025 codeslides
- Reverse Markov Learning: Multi-Step Generative Models for Complex Distributions X. Shen, N. Meinshausen, T. Zhang Preprint, 2025
- Generalized Engression Models X. Shen, Z. Guo, F. Bach Preprint, 2026 code
- Distributional Principal Autoencoders X. Shen, N. Meinshausen Preprint, 2024 codeslides
- Distributional Instrumental Variable Method A. Holovchak, S. Saengkyongam, N. Meinshausen, X. Shen Preprint, 2025 code
- Frugal, Flexible, Faithful: Causal Data Simulation via Frengression L. Yang, J. M. Tarp, R. J. Evans, X. Shen Preprint, 2025 code
- Invariant Probabilistic Prediction A. Henzi, X. Shen, M. Law, P. Bühlmann Biometrika, 2025 code
- Causality-Oriented Robustness: Exploiting General Noise Interventions X. Shen, P. Bühlmann, A. Taeb Journal of the American Statistical Association, 2026 arXivcodeslides
- Sufficiently Reduced Distributional Regression A. Henzi, T. Liu, X. Shen Preprint, 2026 code
- Simulation-Based Empirical Bayes X. Shen, D. Cai, C. Zhang, D. M. Blei Preprint, 2026
Software
- engression — R and Python packages for engression; the Python package also implements generalized engression models. CRAN · PyPI · GitHub
- DistributionalPrincipalAutoencoder — Python package for distributional principal autoencoders. PyPI · GitHub
- DistributionIV — Python package for the distributional instrumental variable method. PyPI
- frengression — Python code for causal data simulation. GitHub
Code for other papers is linked from each entry under Publications.
Selected talks
- Generative Climate Modeling: Emulation and Dimension Reduction — UW Atmospheric and Climate Sciences, Jan 2026 · slides
- Generalization Beyond Observations: A Distributional Perspective — Stanford, Oct 2025 · slides
- Distributional Causal Inference: from Estimation to Simulation — UW, Oct 2025 · slides
- Distributional learning: an overview of my current work — poster
Teaching
University of Washington
- STAT 591: Distributional Learning and Generative Modeling
- STAT 538: Statistical Learning: Modeling, Prediction, and Computing
- STAT 435: Introduction to Statistical Machine Learning
Earlier
- ETH Zürich: AI Center Projects in Machine Learning Research
- HKUST: teaching assistant for Regression Analysis, Statistical Inference, Survival Analysis and Calculus
Service & awards
Service
- Area Chair, ICML 2026 and NeurIPS 2026
- Co-organizer, IMS Workshop on Frontiers in Statistical Machine Learning, 2026
- Co-organizer, Young Data Science Researcher Seminar Zürich, 2022–2023
- Workflow Chair, ICML 2021
- Reviewer for JRSSB, JASA, AoS, JMLR, NeurIPS, ICML, ICLR, UAI and others
Awards
- Royalty Research Fund Award, University of Washington, 2026
- Foundations of Data Science Postdoctoral Fellowship, ETH Zürich, 2023–2025
- ETH AI Center Postdoctoral Fellowship, 2022–2023
- UAI Scholarship, 2022
- Din-Yu Hsieh Teaching Award, HKUST, 2019–2021
Miscellanea
I love the outdoors and enjoy hiking, cycling, tennis and ping pong. I played on the HKUST and Fudan University table tennis teams and was a member of the Fudan University Chinese Orchestra.
