Avatar
Zonghao (Hudson) Chen
Ph.D. candidate, University College London
zonghao.chen.22@ucl.ac.uk

About me

I am a Ph.D. candidate at the Center of Foundational Artificial Intelligence of University College London, supervised by François-Xavier Briol and Arthur Gretton. I am interested in kernel methods with application to generative models, numerical integration and causal inference. Prior to my PhD, I obtained my bachelor's degree from the department of Electronic Engineering, Tsinghua University, 2022.
I organize the JumpTrading/ELLIS CSML seminar series on Computational Statistics and Machine Learning for the UCL ELLIS Unit. Feel free to drop me an email if you would like to give a talk!

Recent News

June 2025


May 2025
I will visit Taiji Suzuki at RIKEN AIP and University of Tokyo from June 2025 to August 2025.

New paper: MMD gradient flow can generate samples with `super-convergence' cubature rate. The first discrete-time finite-particle convergence result for MMD gradient flow.

Journal Publications and Preprints

  • (De)-regularized Maximum Mean Discrepancy Gradient Flow
    Zonghao Chen, Aratrika Mustafi, Pierre Glaser, Anna Korba, Arthur Gretton, Bharath K. Sriperumbudur
    Preprint
    Paper, Code

    • Conference Publications and Preprints

  • Stationary MMD Points for Cubature
    Zonghao Chen, Toni Karvonen, Heishiro Kanagawa, François-Xavier Briol, Chris. J. Oates
    Preprint
    Paper, Code

  • Nested Expectations with Kernel Quadrature
    Zonghao Chen, Masha Naslidnyk, François-Xavier Briol
    ICML '25 International Conference on Machine Learning
    Paper, Code, Poster

  • Conditional Bayesian Quadrature
    Zonghao Chen*, Masha Naslidnyk*, Arthur Gretton, François-Xavier Briol
    UAI '24 Conference on Uncertainty in Artificial Intelligence
    Paper, Code, Poster

  • Conformal Counterfactual Inference under Hidden Confounding
    Zonghao Chen*, Ruocheng Guo*, Jean-François Ton, Yang Liu
    KDD '24 Conference on Knowledge Discovery and Data Mining
    Paper, Code, Poster

  • Tractable Function-Space Variational Inference in Bayesian Neural Networks
    Tim G. J. Rudner, Zonghao Chen, Yee Whye Teh, Yarin Gal
    NeurIPS '22 Conference on Neural Information Processing Systems.
    Paper

  • Efficient Neural Network Training via Forward and Backward Propagation Sparsification
    Xiao Zhou*, Weizhong Zhang*, Zonghao Chen, Shizhe Diao, Tong Zhang
    NeurIPS ’21 Conference on Neural Information Processing Systems
    Paper

  • A Neural Tangent Kernel Perspective on Function-Space Regularization in Neural Networks
    Zonghao Chen*, Xupeng Shi*, Tim G. J. Rudner, Qixuan Feng, Weizhong Zhang, Tong Zhang
    NeurIPS 2022 Workshop: Optimization for machine learning
    Paper

    • Awards

    • RIKEN Visiting Fellowship in Machine Learning, RIKEN AIP, Japan

    • Newcastle Visiting Fellowship in Machine Learning, Newcastle University

    • Tsinghua Presidential Scholarship, Tsinghua University

    • Yinghua Scholarship, Tsinghua University

    Valar Morghulis! Valar Dohaeris!