Matthias Katzfuß

I am a Professor in the Department of Statistics at University of Wisconsin–Madison.

I'm a Fellow of the American Statistical Association, and the recipient of an NSF Career Award, a Fulbright Scholarship, and an Early Investigator Award by the ASA Section on Statistics and the Environment.

Recent group news:

• Postdoc position available.

• 3/2025: Paper on "Sparse inverse Cholesky factorization of dense kernel matrices by greedy conditional selection" accepted at SIAM/ASA Journal on Uncertainty Quantification.

• 2/2025: Paper on "Learning non-Gaussian spatial distributions via Bayesian transport maps with parametric shrinkage" accepted at JABES.

• 1/2025: Paper on "Vecchia Gaussian process ensembles on internal representations of deep neural networks" accepted at AISTATS.

• 1/2025: Paper on "Probabilistic skip connections for deterministic uncertainty quantification in deep neural networks" now on arXiv.

• 12/2024: New NASA grant awarded on Probabilistic Calibration and Data Assimilation for Geophysical Models.

• 11/2024: Paper on "Nonstationary spatial modeling of massive global satellite data" accepted at JCGS.

• 9/2024: Paper on "Learning non-Gaussian spatial distributions via Bayesian transport maps with parametric shrinkage" now on arXiv.

• 8/2024: Anirban has completed his PhD. Congratulations!

• 8/2024: Matthias was named Fellow of the American Statistical Association.

• 7/2024: Matthias has joined the Board of Directors of the ASA Wisconsin Chapter.

• 6/2024: Paper on "Scalable sampling of truncated multivariate normals using sequential nearest-neighbor approximation" now on arXiv.

• 1/2024: Paper on "Asymptotic properties of Vecchia approximation for Gaussian processes" now on arXiv.

• 11/2023: Paper on "Linear-cost Vecchia approximation of multivariate normal probabilities" now on arXiv.

• 10/2023: Paper on "Bayesian nonparametric generative modeling of large multivariate non-Gaussian spatial fields" accepted at JABES.

• 9/2023: Paper on "Locally anisotropic nonstationary covariance functions on the sphere" accepted at JABES.

• 8/2023: Matthias has joined the Department of Statistics at UW–Madison.

• 7/2023: MJ has completed his PhD. Congratulations!

• 5/2023: Paper on "Vecchia Gaussian process ensembles on internal representations of deep neural networks" now on arXiv.

• 4/2023: Paper on "Variational sparse inverse Cholesky approximation for latent Gaussian processes via double Kullback-Leibler minimization" accepted at ICML.

• 3/2023: Paper on "Scalable Bayesian transport maps for high-dimensional non-Gaussian spatial fields" accepted at JASA T&M.

• 3/2023: Paper on "Correlation-based sparse inverse Cholesky factorization for fast Gaussian-process inference" accepted at Statistics and Computing.

• 1/2023: Paper on "Variational sparse inverse Cholesky approximation for latent Gaussian processes via double Kullback-Leibler minimization" now on arXiv.

• 1/2023: Paper on "Scalable Bayesian optimization using Vecchia approximations of Gaussian processes" accepted at AISTATS.

• 10/2022: Paul Wiemann is joining the group as a postdoc.

• 9/2022: Paper on "Scalable Gaussian-process regression and variable selection using Vecchia approximations" accepted at Journal of Machine Learning Research.

• 9/2022: Paper on "High-dimensional nonlinear spatio-temporal filtering by compressing hierarchical sparse Cholesky factors" accepted at Journal of Data Science. 

• 9/2022: Paper on "Ordered conditional approximation of Potts models" accepted at Spatial Statistics.

• 9/2022: Matthias has been promoted to (full) professor.

• 8/2022: Paper on "Locally anisotropic covariance functions on the sphere" now on arXiv.

• 7/2022: Paper on "Scalable spatio-temporal smoothing via hierarchical sparse Cholesky decomposition" accepted at Environmetrics.

• 6/2022: Brian has completed his PhD. Congratulations!

• 5/2022: We are looking for a postdoc to be funded by a newly awarded NASA grant.   The position has been filled.

• 3/2022: Paper on "Scalable Bayesian optimization using Vecchia approximations of Gaussian processes" now on arXiv.

• 2/2022: Paper on "Scalable Gaussian-process regression and variable selection using Vecchia approximations" now on arXiv.

• 1/2022: Matthias is chair of the ISBA Program Council for 2022.

Older news items can be found here.