Shizhou Xu

Welcome

Hello, I’m Shizhou Xu, a postdoctoral researcher at the University of California, Davis. My research focuses on the interdisciplinary area between mathematics and machine learning, particularly in probability theory, optimal transport, and trustworthy AI.

My recent work focuses on the following areas: AI trustworthiness (privacy, unlearning, fairness, robustness), neural optimal transport, multi-modal learning, statistical inference, and LLMs (temperature control and unlearning).

I am actively seeking opportunities where I can continue to contribute to advancing machine learning through rigorous mathematical frameworks.

Research Interests

• Mathematics: Probability theory, optimal transport, stochastic dynamical systems, ergodic theory, and the mathematics of data science.
• Statistics: High-dimensional statistics, mathematical statistics, and statistical inference.
• Machine Learning: Fairness, privacy, robustness, explainable AI, and machine unlearning.

You can read more about my research here.

Recent Publications

Here are a few highlights from my recent work:

• Machine Unlearning via Information Theoretic Regularization, arXiv, 2025.
• WHOMP: Optimizing Randomized Controlled Trials via Wasserstein Homogeneity, arXiv, 2024.
• On the (In)Compatibility between Individual and Group Fairness, arXiv, 2024.
• Fair Data Representation for Machine Learning at the Pareto Frontier, Journal of Machine Learning Research, 2023.

For a full list of my publications, see Publications.

News and Updates

• Mar 2025: To present “WHOMP: Wasserstein Homogeneity Partition” at the Stanford University Conference on Parsimony and Learning (CPAL).
• Jan 2025: Invited talk on “WHOMP: Wasserstein Homogeneity Partition” at the University of California Davis MADDD Seminer.
• Sep 2024: Invited talk on Fairness in Machine Learning at the Computational Harmonic Analysis in Data Science and Machine Learning workshop, Casa Matemática Oaxaca & Banff International Research Station.
• Jul 2024: Presented “Fair Data Representation at the Pareto Frontier” at the International Conference on Machine Learning (ICML).
• Jun 2024: Awarded the Yueh-Jing Lin Scholarship for excellence in research at UC Davis.
• Aug 2023: “Fair Data Representation at the Pareto Frontier” is published in the Journal of Machine Learning Research on fairness in machine learning at the Pareto frontier.

Get in Touch

I am always open to discussing potential collaborations and research opportunities. Feel free to reach out via email or connect with me on LinkedIn.

For more information, you can download my CV or check out my Google Scholar profile.