Welcome to the Github page of the Department of Stochastic Simulation and Safety Research for Hydrosystems (LS3). We are part of the Institute for Modelling Hydraulic and Environmental Systems (IWS) of the Faculty for Civil and Environmental Engineering and the Cluster of Excellence SimTech of the University of Stuttgart. Furthermore, we participate in the Collaborative Research Centers Interface-Driven Multi-Field Processes in Porous Media - Flow, Transport and Deformation.
We are a team of researchers with many different scientific backgrounds from all over the world. This diversity and our great team spirit are essential for excellent research and teaching.
We collaborate strongly with internationally renowned researchers. These collaborations comprise projects related to environmental modelling, underground contaminant transport, statistics, optimization, quantification of uncertainty, system reliability, system security and energy systems.
Our group developed various software tools and resources to aid in our research endeavors. Here are some of our published key projects (including important collaborations):
- FINN: FInite volume Neural Network (FINN) | See 1, 2
- PDEBench: A diverse and comprehensive set of benchmarks for scientific machine learning, including challenging and realistic physical problems | See 3
- DaPC NN: DaPC NN: Deep Arbitrary Polynomial Chaos Neural Network | See 4
- aMR-PC: Arbitrary Multi-Resolution Polynomial Chaos python toolbox | See 5
- aPC Matlab Toolbox: Data-driven Arbitrary Polynomial Chaos: Data-driven Arbitrary Polynomial Chaos Expansion for Machine Learning, Uncertainty quantification and Global sensitivity analysis | See 6, 7, 8
- Gaussian active learning on multi-resolution arbitrary polynomial chaos emulator: Code for the aMR-PC toolbox by Ilja Kröker in the version used for the code in GALMAP_code
- BAL-GPE Matlab Toolbox: BAL-GPE Matlab Toolbox: Bayesian Active Learning for GPE
- BaPC Matlab Toolbox: Bayesian Arbitrary Polynomial Chaos | See 9
- bali: A python package for Bayesian likelihood estimation | See 10
- bayesvalidrox: An open-source, object-oriented Python package for surrogate-assisted Bayesian Validation of computational models. | See 11
- SpectralToolbox: A FFT-based kriging package | See 12
Our research findings and contributions have been published in reputable journals, conferences, and other platforms. You find a list of our publications here
Some notable recent publications:
- Horuz CC, Karlbauer M, Praditia T, Butz MV, Oladyshkin S, Nowak W, et al. Physical Domain Reconstruction with Finite Volume Neural Networks. Applied Artificial Intelligence. 2023;37(1):2204261.
- Oladyshkin S, Praditia T, Kroeker I, Mohammadi F, Nowak W, Otte S. The Deep Arbitrary Polynomial Chaos Neural Network or how Deep Artificial Neural Networks could benefit from Data-Driven Homogeneous Chaos Theory. Neural Networks. 2023;166:85–104.
- Morales Oreamuno MF, Oladyshkin S, Nowak W. Information-Theoretic Scores for Bayesian Model Selection and Similarity Analysis: Concept and Application to a Groundwater Problem. Water Resources Research. 2023 Jul;59(7):e2022WR033711.
- Bürkner PC, Kröker I, Oladyshkin S, Nowak W. The sparse Polynomial Chaos expansion: a fully Bayesian approach with joint priors on the coefficients and global selection of terms. Journal of Computational Physics. 2023;112210.
Interested in collaborating or joining our team? We welcome researchers, developers, and enthusiasts passionate about hydrology, bayesian methods and machine learning. Feel free to contact us for opportunities or open issues and pull requests in our repositories.
We look forward to connecting with you and advancing our research together!