GLUEd Geometric Learning & Uncertainty in Edinburgh

Institute for Adaptive and Neural Computation · School of Informatics · University of Edinburgh

Primary Research Directions

  • Geometric learning: exploiting the structure of data (e.g., symmetries), learning on structured domains (e.g., graphs, manifolds) (to achieve data efficiency and respect constraints)
  • Uncertainty Quantification: building models that quantify uncertainty associated with their own predictions (to support model-based decision-making and safety-critical applications)

We work on geometric learning and uncertainty quantification—separately, and where they meet.

Where these directions meet
Future: Leveraging implicit structure of data to build powerful uncertainty-aware models for complex, high-dimensional problems.
To date: Geometric Gaussian processes—probabilistic models operating on non-Euclidean domains such as manifolds and graphs.
Additional Research Directions: Gaussian Processes Machine Learning Theory AI for Science

Interested in joining? See vacancies. Note: A fully-funded PhD call is open for applications until 31st December 2025 (or until filled).

Members

Former Students

Colin Doumont
(2024, ETH Zurich)

Master's Thesis: Accepted as NeurIPS paper

Next Step: PhD, University of Tübingen

Kacper Wyrwał
(2024, ETH Zurich and University of Edinburgh)

Bachelor's Thesis: Published as ICLR paper (oral)

Next Step: MSc, University of Oxford

Iskander Azangulov
(2022, St. Petersburg University)

Master's Thesis: Published as two JMLR papers: 1st, 2nd

Next Step: PhD, University of Oxford