Ilja Kröker

Dr. Ilja Kröker

orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 February 2011 in Applied Numerical Mathematics
Verified
Journal Article
Closed access
IKIlja Kröker
Ilja Kröker
CRC. Rohde
C. Rohde

9 November 2018 in Computational Geosciences
Verified
Journal Article
Closed access
PDF file
MKMarkus Köppel
Markus Köppel
FFFabian Franzelin
Fabian Franzelin
IKIlja Kröker
Ilja Kröker
SOSergey Oladyshkin
Sergey Oladyshkin
GSGabriele Santin
Gabriele Santin
CRChristian Rohde
Christian Rohde

13 August 2020 in Entropy
Verified
Journal Article
Open access
PDF file
SOSergey Oladyshkin
Sergey Oladyshkin
FMFarid Mohammadi
Farid Mohammadi
IKIlja Kröker
Ilja Kröker
WNWolfgang Nowak
Wolfgang Nowak
Gaussian process emulators (GPE) are a machine learning approach that replicates computational demanding models using training runs of that model. Constructing such a surrogate is very challenging and, in the context of Bayesian inference, the training runs should be well invested. The current paper offers a fully Bayesian view on GPEs for Bayesian inference accompanied by Bayesian active learning (BAL). We introduce three BAL strategies that adaptively identify training sets for the GPE using information-theoretic arguments. The first strategy relies on Bayesian model evidence that indicates the GPE’s quality of matching the measurement data, the second strategy is based on relative entropy that indicates the relative information gain for the GPE, and the third is founded on information entropy that indicates the missing information in the GPE. We illustrate the performance of our three strategies using analytical- and carbon-dioxide benchmarks. The paper shows evidence of convergence against a reference solution and demonstrates quantification of post-calibration uncertainty by comparing the introduced three strategies. We conclude that Bayesian model evidence-based and relative entropy-based strategies outperform the entropy-based strategy because the latter can be misleading during the BAL. The relative entropy-based strategy demonstrates superior performance to the Bayesian model evidence-based strategy.

29 July 2013 in ZAMM - Journal of Applied Mathematics and Mechanics / Zeitschrift für Angewandte Mathematik und Mechanik
Verified
Journal Article
Open access
PDF file
RBRaimund Bürger
Raimund Bürger
IKIlja Kröker
Ilja Kröker
CRC. Rohde
C. Rohde
The continuous sedimentation process in a clarifier‐thickener can be described by a scalar nonlinear conservation law for the local solids volume fraction. The flux density function is discontinuous with respect to spatial position due to feed and discharge mechanisms. Typically, the feed flow cannot be given deterministically and efficient numerical simulation requires a concept for quantifying uncertainty. In this paper uncertainty quantification is expressed by a new hybrid stochastic Galerkin (HSG) method that extends the classical polynomial chaos approximation by multiresolution discretization in the stochastic space. The new approach leads to a deterministic hyperbolic system for a finite number of stochastic moments which is however partially decoupled and thus allows efficient parallelisation. The complexity of the problem is further reduced by stochastic adaptivity. For the approximate solution of the resulting high‐dimensional system a finite volume scheme is introduced. Numerical experiments cover one‐ and two‐dimensional situations.

1 June 2016 in Computers & Chemical Engineering
Verified
Journal Article
Closed access
ABAndrea Barth
Andrea Barth
RBRaimund Bürger
Raimund Bürger
IKIlja Kröker
Ilja Kröker
CRChristian Rohde
Christian Rohde

10 January 2015 in Computational Geosciences
Verified
Journal Article
Closed access
PDF file
IKIlja Kröker
Ilja Kröker
WNWolfgang Nowak
Wolfgang Nowak
CRChristian Rohde
Christian Rohde