Lecturer (Assistant Prof.) in Applied Mathematics at the University of Glasgow. Member of ELLIS Society. Current.

Postdoctoral Researcher Computer Science, Mila - Quebec AI Institute & McGill University. 2022-2023.

Ph.D. Mathematics, University of Edinburgh (MIGSAA CDT). 2022.

Turing Enrichment student, Alan Turing Institute. 2021.

M. Sc. Theoretical Physics, Perimeter Scholars International Award, Perimeter Institute for Theoretical Physics and the University of Waterloo. 2016.

M. Sc. Applied Geophysics (Cum Laude), IDEA League Scholarship, Delft University of Technology, ETH Zürich, and RWTH Aachen. 2015.
Research Intern, Schlumberger Gould Research. 2015.

B. Sc. Physics, Leiden University. 2013.
Assistant Junior Science Lab, Leiden University. 2013.

Leadership Experience and Development:

Teaching Experience:

  • Mathematics 1G: Introduction to Algebra, Geometry & Networks (MATHS1016), University of Glasgow, 2024.
  • Tutor, University of Edinburgh, courses: Machine Learning in Python, Computing and Numerics (Python), Probability with Applications, Several Variable Calculus and Differential Equations, Engineering Mathematics, Probability, 2018-2022.
  • Tutor, Leiden University, covered all courses of the 1st year Physics BSc, 2012-2014.

Service:

  • Area Chair, ICLR “Mathematical and Empirical Understanding of Foundation Models” workshop, 2023.
  • Reviewer for Advances in Approximate Bayesian Inference (AABI), 2023.
  • Reviewer for CVPR, Women in Computer Vision Workshop, 2023.
  • Reviewer for ICML, 2021 and 2022.
  • Reviewer for the NeurIPS “Machine Learning and the Physical Sciences” and “ML Retrospectives, Surveys and Meta-Analyses” workshops, 2020.
  • Member of Mathematics Outreach Team, University of Edinburgh, 2018.
  • MIGSAA Mentor, 2018.
  • Machine Learning Reading Group Organizer, Mathematics Department, University of Edinburgh, 2018.
  • MIGSAA Student Representative for Scottish Mathematical Sciences Training Centre, 2017.

Software: Python, PyTorch, MATLAB.

Publications:

  • Müller, M., Vlaar, T., Rolnick, D., and Hein, M., “Normalization Layers Are All That Sharpness-Aware Minimization Needs”, NeurIPS 2023. Available: arXiv:2306.04226
  • Vlaar, T. and Leimkuhler, B., “Multirate Training of Neural Networks”, ICML, PMLR 162, 2022. Available: arXiv:2106.10771
  • Vlaar, T. and Frankle, J., “What can linear interpolation of neural network loss landscapes tell us?”, ICML, PMLR 162, 2022. Available: arXiv:2106.16004
  • Vlaar, T., “Neural Network Behavior at the Classification Boundary”, poster accepted for NeurIPS 2021 I (Still) Can’t Believe It’s Not Better (ICBINB) workshop. Preprint available on request.
  • Leimkuhler, B., Vlaar, T., Pouchon, T., and Storkey, A., “Better Training using Weight-Constrained Stochastic Dynamics”, ICML, PMLR 139, 2021. Available: arXiv:2106.10704
  • Leimkuhler, B., Pouchon, T., Vlaar, T., and Storkey, A., “Constraint-Based Regularization of Neural Networks” (2020). Accepted as spotlight presentation for NeurIPS 2020 Optimization for ML workshop and received best student paper award. Available arXiv:2006.10114
  • Leimkuhler, B., Matthews, C., and Vlaar, T.J., “Partitioned integrators for thermodynamic parameterization of neural networks”, Foundations of Data Science 1 (4) , 457-489 (2019). Accepted as a digital acceptance to NeurIPS 2019 Machine Learning and the Physical Sciences workshop. Available arXiv:1908.11843
  • Chojnacki, L., Cook, C.Q., Dalidovich, D., Hayward Sierens, L.E., Lantagne-Hurtubise, É., Melko, R.G., and Vlaar, T.J., “Shape dependence of two-cylinder Rényi entropies for free bosons on a lattice”, Physical Review B 94, 165136 (2016). Available arXiv:1607.05311.