• Harrison Zhu, Adam Howes, Owen van Eer, Maxime Rischard, Dino Sejdinovic and Seth Flaxman (2021).Multi-resolution Spatial Regression for Aggregated Data with an Application to Crop Yield Prediction. ArXiv. [Manuscript] [Code]


  • Xing Liu, Harrison Zhu, Jean-François Ton, George Wynne, Andrew Duncan (2022). Grassmann Stein Variational Gradient Descent. To appear in the 25th International Conference on Artificial Intelligence and Statistics. [Manuscript] [Code]
  • Harrison Zhu, Xing Liu, Ruya Kang, Zhichao Shen, Seth Flaxman and François-Xavier Briol (2020). Bayesian probabilistic numerical integral with tree based priors. 34th Conference on Advances in Neural Information Processing Systems 2020. [Manuscript] [Code]
  • Seth Flaxman, Swapnil Mishra, Axel Gandy, H Juliette T Unwin, Thomas A Mellan, Helen Coupland, Charles Whittaker, Harrison Zhu, Tresnia Berah, Jeffrey W Eaton, Mélodie Monod, Azra C Ghani, Christl A Donnelly, Steven M Riley, Michaela AC Vollmer, Neil M Ferguson, Lucy C Okell, Samir Bhatt (2020). Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe. Nature pp1-8. [Manuscript] [Code]


  • Harrison Zhu, Xing Liu, Alberto Caron, Ioanna Manolopoulou, Seth Flaxman and François-Xavier Briol (2020). Contributed Discussion of “Bayesian Regression Tree Models for Causal Inference: Regularization, Confounding, and Heterogeneous Effects”. Bayesian Analysis, 15(3), to appear. [Journal (pp55-58)] [Manuscript Code]

Talks and Presentations

  • Sparse Variational Gaussian Processes - 12 November 2021. CSML Reading Group, Imperial College London. [Slides] [Demo]
  • Gaussian Processes - 06 February 2019, Huxley 213 Clore lecture Theatre, Imperial College London. [Slides]
  • Urysohn Lemma and its Applications - 24 October 2018, Huxley 213 Clore Lecture Theatre, Imperial College London. [Slides]

Notes and Extracts

Here are some random notes that I thought would be useful for a casual read. A lot of them contain mistakes and so please read them with your own caution.


Side Projects