Han Cheng Lie (pronouns: he / him)
Professor (W1), Chair of Uncertainty Quantification, Institut für Mathematik, Universität Potsdam
Very important science
NEW: 2023 AR6 Synthesis Report. A key figure that illustrates where action can be taken is here. For more figures, see this page.
The evidence for the climate crisis (see also the Sixth Assessment Report of the IPCC)
What individuals can do about the climate crisis [Link to the cited research article]
2019 Global Assessment Report on Biodiversity and Ecosystem Services
2019 IPCC Special Report on Global Warming of 1.5 degrees Celcius
Updates
November 2024:
The first International Conference on Probabilistic Numerics (ProbNum25) will take place 1 through 3 September 2025 at EURECOM in Sophia Antipolis near Nice, France. The conference provides a venue for technically and mathematically ambitious research in Probabilistic Numerics, and a community of researchers who are interested in the quantification of computational uncertainty. Contributed work and archival proceedings will be published as a volume of the Proceedings of Machine Learning Research. The organisers are Motonobu Kanagawa, Alexandra Gessner, Jon Cockayne, and Philipp Hennig.
A preprint on low-rank approximations of posteriors for linear Gaussian inverse problems on Hilbert spaces, co-authored with Giuseppe Carere, appears on arXiv.
September 2024:
A paper on data subsampling for Poisson regression, co-authored with Alex Munteanu, is accepted for publication in the proceedings of NeurIPS. The arXiv version is here.
August 2024:
A preprint on generalised rank-constrained approximations of Hilbert-Schmidt operators, co-authored with Giuseppe Carere, appears on arXiv.
Giuseppe Carere and I organised a contributed paper session on approximate Bayesian inference at the 11th Bernoulli-IMS World Congress in Bochum.
July 2024:
A preprint on Bayesian nonparametric inference for covariate-parameter relationships in population modelling and pharmacology appears on arXiv.
The paper on choosing observation operators to mitigate model error in Bayesian inverse problems (joint with Nada Cvetkovic, Harshit Bansal, and Karen Veroy from Eindhoven University of Technology) is published online at the SIAM/ASA Journal of Uncertainty Quantification.
Archived updates are available here.