Search Results for author: Matthias Katzfuss

Found 10 papers, 8 papers with code

Scalable Bayesian Optimization Using Vecchia Approximations of Gaussian Processes

no code implementations2 Mar 2022 Felix Jimenez, Matthias Katzfuss

We focus on the use of our warped Vecchia GP in trust-region Bayesian optimization via Thompson sampling.

Gaussian Processes

Scalable Gaussian-process regression and variable selection using Vecchia approximations

1 code implementation25 Feb 2022 Jian Cao, Joseph Guinness, Marc G. Genton, Matthias Katzfuss

Gaussian process (GP) regression is a flexible, nonparametric approach to regression that naturally quantifies uncertainty.

Variable Selection

Scalable penalized spatiotemporal land-use regression for ground-level nitrogen dioxide

1 code implementation19 May 2020 Kyle P Messier, Matthias Katzfuss

Nitrogen dioxide (NO$_2$) is a primary constituent of traffic-related air pollution and has well established harmful environmental and human-health impacts.

Applications

Scaled Vecchia approximation for fast computer-model emulation

1 code implementation1 May 2020 Matthias Katzfuss, Joseph Guinness, Earl Lawrence

Many scientific phenomena are studied using computer experiments consisting of multiple runs of a computer model while varying the input settings.

Gaussian Processes

Sparse Cholesky factorization by Kullback-Leibler minimization

1 code implementation29 Apr 2020 Florian Schäfer, Matthias Katzfuss, Houman Owhadi

We propose to compute a sparse approximate inverse Cholesky factor $L$ of a dense covariance matrix $\Theta$ by minimizing the Kullback-Leibler divergence between the Gaussian distributions $\mathcal{N}(0, \Theta)$ and $\mathcal{N}(0, L^{-\top} L^{-1})$, subject to a sparsity constraint.

Numerical Analysis Numerical Analysis Optimization and Control Statistics Theory Computation Statistics Theory

Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data

2 code implementations18 Jun 2019 Daniel Zilber, Matthias Katzfuss

Generalized Gaussian processes (GGPs) are highly flexible models that combine latent GPs with potentially non-Gaussian likelihoods from the exponential family.

Methodology Computation

Vecchia approximations of Gaussian-process predictions

1 code implementation8 May 2018 Matthias Katzfuss, Joseph Guinness, Wenlong Gong

Gaussian processes (GPs) are highly flexible function estimators used for geospatial analysis, nonparametric regression, and machine learning, but they are computationally infeasible for large datasets.

Methodology Computation

A general framework for Vecchia approximations of Gaussian processes

1 code implementation21 Aug 2017 Matthias Katzfuss, Joseph Guinness

Gaussian processes (GPs) are commonly used as models for functions, time series, and spatial fields, but they are computationally infeasible for large datasets.

Methodology Computation

A multi-resolution approximation for massive spatial datasets

3 code implementations16 Jul 2015 Matthias Katzfuss

The M-RA process is specified as a linear combination of basis functions at multiple levels of spatial resolution, which can capture spatial structure from very fine to very large scales.

Methodology Computation

Bayesian Nonstationary Spatial Modeling for Very Large Datasets

no code implementations10 Apr 2012 Matthias Katzfuss

With the proliferation of modern high-resolution measuring instruments mounted on satellites, planes, ground-based vehicles and monitoring stations, a need has arisen for statistical methods suitable for the analysis of large spatial datasets observed on large spatial domains.

Methodology Applications Computation

Cannot find the paper you are looking for? You can Submit a new open access paper.