October 2025: Presenting a poster on the TMLR paper 'Numerically Robust Fixed-Point Smoothing Without State Augmentation' at the ELLIS UnConference in Copenhagen
October 2025: Preprint 'Matrix-Free Least Squares Solvers: Values, Gradients, and What to Do With Them' posted on arXiv
September 2025: Paper on approximate Bayesian neural operators accepted and published by TMLR
September 2025: Presented 'Adaptive probabilistic ODE solvers without adaptive memory requirements' at the International Conference on Probabilistic Numerics in Nice; demoed Probdiffeq (a JAX library)
June 2025: Paper on adaptive probabilistic ODE solvers accepted at ICPN
March 2025: Preprint 'Numerically robust Gaussian state estimation with singular observation noise' posted on arXiv
January 2025: Paper 'Numerically robust fixed-point smoothing without state augmentation' published by TMLR
December 2024: Speaking at the D3S3 Workshop at NeurIPS 2024
November 2024: Gave talks on adaptive ODE solvers in Oxford, Cambridge, and at DTU Compute’s CUQI seminar series
October 2024: Paper 'Gradients of functions of large matrices' accepted as a spotlight at NeurIPS
October 2024: Preprint 'Adaptive probabilistic ODE solvers without adaptive memory requirements' uploaded to arXiv
Background
Machine learning researcher based in Copenhagen
Works on AI4Science at the intersection of numerical methods and machine learning
Research interests include probabilistic numerics, differentiable programming, state-space models, numerical linear algebra, Bayesian machine learning, and physics-informed machine learning
Primarily codes in Python/JAX
Member of the ELLIS Society and affiliated researcher at the Pioneer Centre for AI (P1)
Goes by 'Nico' informally but uses 'Nicholas' in academic publications