About

My name is Kaiyu Li, a PhD student in Statistics at University College London. I am supervised by Prof. Serge Guillas and Prof. F-X Briol.

Research Interest

I am interested in uncertainty quantification, computational statistics and machine learning. My work involves developing and applying new methods to investigate and quantify uncertainties in computer models.

Publication

• Li, K., ; Yang, Y.; Chen, X.; He, Y.; Sun, Z. (2026). Multilevel Control Functional. In International Conference on Learning Representations (ICLR 2026). Score: 8/8/8 ⭐ Rank 2nd / 19,000 submissions
• Cui, Y., Song, Y., Feng, K., Xu, H., Wei, Q., & Li, K.. (2025). Chance-constrained optimal power flow for improving line flow and voltage security of power transmission networks. Autonomous Intelligent Systems, 5(1), 31.
• Li, K., Salmanidou, D., Gopinathan, D., Heidarzadeh, M., & Guillas, S. (2025). Uncertainty in Manning’s roughness coefficient in multilevel simulations of future tsunamis in Sumatra. Proceedings of the Royal Society A, 481(2316), p.20240637.
• Li, K., Giles, D., Karvonen, T., Guillas, S., & Briol, F. X. (2023). Multilevel Bayesian Quadrature. In Proc. of International Conference on Artificial Intelligence and Statistics (AISTATS 2023). PMLR. ⭐ Oral presentation (Top 6% of accepted papers)

Talk/Presentation

• Poster at LSIT 2023 - Seventh London Symposium on Information Theory, May 2023, UK.
• Oral presentation & Poster at The 26th International Conference on Artificial Intelligence and Statistics (AISTATS), April 2023, Spain.
• Poster at IMSI Workshop - Expressing and Exploiting Structure in Modeling, Theory, and Computation with Gaussian Processes, August 2022, USA.
• Talk at SIAM Conference on Uncertainty Quantification, April 2022, USA.
• Talk at High Dimensional and Functional Data Group, April 2022, UK.
• Poster at The Machine Learning Summer School by the Max Planck Institute for Intelligent Systems, June 2020, Germany.