Research Interests:
-Basic research in statistical machine learning (theory and methodology)
-Counter-intuitive properties of high-dimensional data spaces
-Probabilistic modelling of multivariate data, Bayesian inference
-Large scale heuristic optimisation
-Dimensionality reduction, random projections, compressive learning, compressive optimisation, model compression

University Homepage - Publications - Google Scholar profile - Projects

NEW: We are organising HDM 24

Note to prospective PhD applicants: I welcome well-qualified and motivated PhD candidates throughout the year. Read this and email me with your transcript and research proposal in pdf format, or apply directly via the portal. If you are looking for ideas to develop your proposal, check out the topics above, including the HDM 24 workshop topics. In addition, below are some broad research question examples that interest me:

- When, why, and how machines can learn to generalize correctly from a finite sample?
- When, why, and how does a random linear projection of a data set preserve norms and distances?
- How can we bridge the gap between theoretical guarantees and practical experiences in machine learning?
- How can we explain that model compression works so well in practice?
- How do high dimensional data spaces look like?