KIPS (Kernels and Information Processing Systems) research group is a subset of the larger Computational Statistics and Machine Learning (OxCSML) network within the Department of Statistics and we closely collaborate with other researchers in OxCSML. Our research spans a variety of topics at the interface between statistical methodology and machine learning, including:

  • Large-scale nonparametric and kernel methods,
  • Multiresolution data and weak supervision,
  • Robust machine learning: robustness to model misspecification, censoring, spatiotemporal confounding,
  • Measures of dependence and multivariate interaction, causal inference,
  • Meta learning, hierarchical and deep generative modelling.

DPhil Students

  • Shahine Bouabid
    kernel methods, Bayesian nonparametrics, deep learning, aerosol-cloud interaction

  • Valerie Bradley
    kernel methods, selection bias

  • Anthony Caterini
    deep generative models, variational inference

  • Siu Lun Chau
    kernel methods, preference learning, causality, explainable AI

  • Robert Hu
    large scale kernel methods, causality, generative models, interpretability

  • Zhu Li
    kernel methods, learning theory

  • David Rindt
    survival analysis, kernel methods

  • Jean-Francois Ton
    kernel methods, meta learning, causality

  • Veit Wild
    Bayesian nonparametrics, Gaussian processes, variational inference


Associate Members

Rob Zinkov


Emiliano Diaz Salas Porras, Oct-Dec 2019