Forschung & Publikationen
Wir nutzten Partnerschaften mit Universitäten, um im dynamischen Feld von Data Science und KI up-to-date zu bleiben. Durch die Fusion von Wissenschaft und Praxis erzielen wir bahnbrechende Fortschritte.
Wandelnden Disziplinen
Für uns ist diese Einbindung in die akademische Welt sehr wertvoll, da wissenschaftliche Fragestellungen häufig auf unternehmerische Kontexte angewendet werden können. Speziell in Disziplinen wie der Medizin kann der Einsatz von KI-Methoden bedeutende Fortschritte ermöglichen. Als Fachleute auf dem Gebiet KI und Data Science haben wir die Möglichkeit, unser Fachwissen auf verschiedene Anwendungsbereiche anzuwenden und dabei zu neuen Erkenntnissen beizutragen.
Ein gutes Beispiel für diese Verbindung von Wissenschaft und Praxis ist unser Mitbegründer, Dr. Mattis Hartwig, der auch weiterhin als Senior Researcher am Deutschen Forschungsinstitut für Künstliche Intelligenz tätig ist. Seine enge Bindung an die akademische Forschung ermöglicht es uns, Theorie und Praxis miteinander zu verbinden und aktuelle wissenschaftliche Erkenntnisse in unsere Arbeit zu integrieren.
Außerdem besteht die Möglichkeit, bei uns eine Abschlussarbeit in Informatik, Data Science und verwandten Gebieten zu absolvieren, weiteres findest du hier.
2026
2025
- Florian Kretzschmar, Hannes Reinhardt, Mattis Hartwig, „Vom ETL-Altbau ins Lakehouse,“
BI Spektrum, , 06, 2025.
@article{Kretzschmar2025-38,
title: "Vom ETL-Altbau ins Lakehouse",
author: "Florian Kretzschmar and Hannes Reinhardt and Mattis Hartwig",
year: "2025"
journal: "BI Spektrum"
pages: ""
}
2024
- Mattis Hartwig, Ralf Möller, Tanya Braun, „An Extended View on Lifting Gaussian Bayesian Networks,“
Artificial Intelligence, vol. 330, pp. 104082, 05, 2024.
@article{Hartwig2024-35,
title: "An Extended View on Lifting Gaussian Bayesian Networks",
author: "Mattis Hartwig and Ralf Möller and Tanya Braun",
year: "2024"
journal: "Artificial Intelligence"
pages: ""
}
2023
- Johanna Ott , Arthur Ledaguenel, Céline Hudelot, Mattis Hartwig, „How to Think About Benchmarking Neurosymbolic AI?,“
in 17th International Workshop on Neural-Symbolic Learning and Reasoning -NESY 2023, 2023, pp. 1-8. PDF ansehen - Melle Mendikowski, Benjamin Schindler, Thomas Schmid, Ralf Möller, and Mattis Hartwig, „Improved Techniques for Training Tabular GANs Using Cramer’s V Statistics,“
in Proceedings of the Canadian Conference on Artificial Intelligence, 2023, pp. 1-12. PDF ansehen - Nils Frohwitter, Alessa Hering, Ralf Möller, Mattis Hartwig, „Evaluating the Effects of a Priori Deep Learning Image Synthesis on Multi-Modal MR-to-CT Image Registration Performance,“
in Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2023), 2023, pp. 322-329. PDF ansehen - Alex Winter, Toralf Kirsten, Mattis Hartwig, „Predicting Hospital Length of Stay of Patients Leaving the Emergency Department,“
in Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2023) – HEALTHINF, 2023, pp. 124-131. PDF ansehen
@conference{2023-45,
title: "How to Think About Benchmarking Neurosymbolic AI?",
author: "Johanna Ott and Arthur Ledaguenel and Céline Hudelot and Mattis Hartwig",
year: "2023"
booktitle: "17th International Workshop on Neural-Symbolic Learning and Reasoning -NESY 2023"
pages: "1 -- 8"
}
@conference{Mendikowski2023-7,
title: "Improved Techniques for Training Tabular GANs Using Cramer’s V Statistics",
author: "Melle Mendikowski and Benjamin Schindler and Thomas Schmid and Ralf Möller and and Mattis Hartwig",
year: "2023"
booktitle: "Proceedings of the Canadian Conference on Artificial Intelligence"
pages: "1 -- 12"
}
@conference{Frohwitter2023-17,
title: "Evaluating the Effects of a Priori Deep Learning Image Synthesis on Multi-Modal MR-to-CT Image Registration Performance",
author: " Nils Frohwitter and Alessa Hering and Ralf Möller and Mattis Hartwig",
year: "2023"
booktitle: "Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2023)"
pages: " 322 -- 329"
}
@conference{Winter2023-25,
title: "Predicting Hospital Length of Stay of Patients Leaving the Emergency Department",
author: "Alex Winter and Toralf Kirsten and Mattis Hartwig",
year: "2023"
booktitle: "Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2023) - HEALTHINF"
pages: "124 -- 131"
}
2022
- Mattis Hartwig, „New Methods for Efficient Query Answering in Gaussian Probabilistic Graphical ModelsInstitut für Informationsssysteme, Universität zu Lübeck, Lübeck, 2022.
- Melle Mendikowski, Mattis Hartwig, „Creating Customers That Never Existed: Synthesis of E-commerce Data Using CTGAN,“
in 18th International Conference on Machine Learning and Data Mining MLDM, 2022, pp. 91-105. PDF ansehen - Moritz P. Hoffmann, Tanya Braun, Ralf Möller, „Lifted Division for Lifted Hugin Belief Propagation,“
in , 2022, pp. 6501-6510. PDF ansehen - Mattis Hartwig, Tanya Braun, Ralf Möller, „Increasing State Estimation Accuracy in the Inference Algorithm on a Hybrid Factor Graph Model,“
in The International FLAIRS Conference Proceedings, vol. 35, 2022. [Online]. Verfügbar: https://doi.org/10.32473/flairs.v35i.130682. PDF ansehen
@phdthesis{Hartwig2022-40,
title: "New Methods for Efficient Query Answering in Gaussian Probabilistic Graphical Models",
author: "Mattis Hartwig",
year: "2022"
}
@conference{Mendikowski2022-30,
title: "Creating Customers That Never Existed: Synthesis of E-commerce Data Using CTGAN",
author: "Melle Mendikowski and Mattis Hartwig",
year: "2022"
booktitle: "18th International Conference on Machine Learning and Data Mining MLDM"
pages: "91 -- 105"
}
@conference{Hoffmann2022-42,
title: "Lifted Division for Lifted Hugin Belief Propagation",
author: "Moritz P. Hoffmann and Tanya Braun and Ralf Möller",
year: "2022"
booktitle: ""
pages: "6501 -- 6510"
}
@conference{Hartwig2022-5,
title: "Increasing State Estimation Accuracy in the Inference Algorithm on a Hybrid Factor Graph Model",
author: "Mattis Hartwig and Tanya Braun and Ralf Möller",
year: "2022"
booktitle: "The International FLAIRS Conference Proceedings"
pages: ""
}
2021
- Mattis Hartwig, Tanya Braun, Ralf Möller, „Handling Overlaps When Lifting Gaussian Bayesian Networks,“
in Thirtieth International Joint Conference on Artificial Intelligence, vol. 30, 2021. [Online]. Verfügbar: https://doi.org/10.24963/ijcai.2021/581. PDF ansehen - Mattis Hartwig, Achim Peters, „Cooperation and Social Rules Emerging From the Principle of Surprise Minimization,“
Frontiers in Psychol., vol. 10, , Januar, 2021. PDF ansehen
@conference{Hartwig2021-29,
title: "Handling Overlaps When Lifting Gaussian Bayesian Networks",
author: "Mattis Hartwig and Tanya Braun and Ralf Möller",
year: "2021"
booktitle: "Thirtieth International Joint Conference on Artificial Intelligence"
pages: ""
}
@article{Hartwig2021-13,
title: "Cooperation and Social Rules Emerging From the Principle of Surprise Minimization",
author: "Mattis Hartwig and Achim Peters",
year: "2021"
journal: "Frontiers in Psychol."
pages: ""
}
2020
- Mattis Hartwig, Rolf Möller, „How to Encode Dynamic Gaussian Bayesian Networks as Gaussian Processes?,“
in Advances in Artificial Intelligence, 2020. [Online]. Verfügbar: http://dx.doi.org/10.1007/978-3-030-64984-5_29. PDF ansehen - Mattis Hartwig, Ralf Möller, „Lifted Query Answering in Gaussian Bayesian Networks,“
in International Conference on Probabilistic Graphical Models, vol. 138, 2020, pp. 233-244. PDF ansehen - Mattis Hartwig, „Efficient Query Answering in Nonparametric Probabilistic Graphical Models,“
in German Conference on Artificial Intelligence, 2020, pp. 1-5. PDF ansehen - Marisa Mohr, Florian Wilhelm, Mattis Hartwig, Ralf Möller, Karsten Keller, „New Approaches in Ordinal Pattern Representations for Multivariate Time Series,“
in International Florida Artificial Intelligence Research Society Conference, 2020, pp. 124-129. PDF ansehen - Mattis Hartwig, Marisa Mohr, Ralf Möller, „Constructing Gaussian Processes for Probabilistic Graphical Models,“
in FLAIRS Conference, 2020, pp. 57-62. PDF ansehen
@conference{Hartwig2020-11,
title: "How to Encode Dynamic Gaussian Bayesian Networks as Gaussian Processes?",
author: "Mattis Hartwig and Rolf Möller",
year: "2020"
booktitle: "Advances in Artificial Intelligence"
pages: ""
}
@conference{Hartwig2020-25,
title: "Lifted Query Answering in Gaussian Bayesian Networks",
author: "Mattis Hartwig and Ralf Möller",
year: "2020"
booktitle: "International Conference on Probabilistic Graphical Models"
pages: "233 -- 244"
}
@conference{Hartwig2020-8,
title: "Efficient Query Answering in Nonparametric Probabilistic Graphical Models",
author: "Mattis Hartwig",
year: "2020"
booktitle: "German Conference on Artificial Intelligence"
pages: "1 -- 5"
}
@conference{Mohr2020-41,
title: "New Approaches in Ordinal Pattern Representations for Multivariate Time Series",
author: "Marisa Mohr and Florian Wilhelm and Mattis Hartwig and Ralf Möller and Karsten Keller",
year: "2020"
booktitle: "International Florida Artificial Intelligence Research Society Conference"
pages: "124 -- 129"
}
@conference{Hartwig2020-3,
title: "Constructing Gaussian Processes for Probabilistic Graphical Models",
author: "Mattis Hartwig and Marisa Mohr and Ralf Möller",
year: "2020"
booktitle: "FLAIRS Conference"
pages: "57 -- 62"
}
2019
- Mattis Hartwig, Marcel Gehrke, Ralf Möller, „Approximate Query Answering in Complex Gaussian Mixture Models,“
in IEEE International Conference on Big Knowledge, 2019, pp. 81-86. PDF ansehen
@conference{Hartwig2019-31,
title: "Approximate Query Answering in Complex Gaussian Mixture Models",
author: "Mattis Hartwig and Marcel Gehrke and Ralf Möller",
year: "2019"
booktitle: "IEEE International Conference on Big Knowledge"
pages: "81 -- 86"
}
Learning from 3D Image Reconstruction: Reconstructing Spatiotemporal Distributions of Radioactive Iodine after Nuclear Accidents
Autor:innen: Max Friedrich , Mareike Böckel, Oliver Meisenberg, Kathrin Meisenberg, Mattis Hartwig
Multi-dependence Conditional Vector Boosting Categorical Fidelity in Tabular-Data GANs
Autor:innen: Melle Mendikowski, Benjamin Schindler, Thomas Schmid, Ralf Möller, Mattis Hartwig
Predicting the Bed Occupancy in a Hospital
Autor:innen: Simon Schiff, Natalie Kohler, Sebastian Wolfrum, Ralf Möller, Mattis Hartwig
Vom ETL-Altbau ins Lakehouse
Autor:innen: Florian Kretzschmar, Hannes Reinhardt, Mattis Hartwig
@article{Kretzschmar2025-38,
title: "Vom ETL-Altbau ins Lakehouse",
author: "Florian Kretzschmar and Hannes Reinhardt and Mattis Hartwig",
year: "2025"
journal: "BI Spektrum"
pages: ""
}
Approximate Lifted Model Construction
Autor:innen: Malte Luttermann, Jan Speller, Marcel Gehrke, Tanya Braun, Ralf Möller, Mattis Hartwig
Aggregating Predicted Individual Hospital Length of Stay to Predict Bed Occupancy for Hospitals
Autor:innen: Mattis Hartwig, Simon Schiff, Sebastiam Wolfrum, Ralf Möller
Towards Privacy-Preserving Relational Data Synthesis via Probabilistic Relational Models
Autor:innen: Malte Luttermann, Ralf Möller, Mattis Hartwig
Using Data Synthesis to Improve Length of Stay Predictions for Patients with Rare Diagnoses
Autor:innen: Simon Schiff, Sebastian Wolfrum, Ralf Möller, Mattis Hartwig
An Extended View on Lifting Gaussian Bayesian Networks
Autor:innen: Mattis Hartwig, Ralf Möller, Tanya Braun
@article{Hartwig2024-35,
title: "An Extended View on Lifting Gaussian Bayesian Networks",
author: "Mattis Hartwig and Ralf Möller and Tanya Braun",
year: "2024"
journal: "Artificial Intelligence"
pages: ""
}
Lifted Causal Inference in Relational Domains
Autor:innen: Malte Luttermann, Mattis Hartwig, Tanya Braun, Ralf Möller, Marcel Gehrke
How to Think About Benchmarking Neurosymbolic AI?
PDF ansehenAutor:innen: Johanna Ott , Arthur Ledaguenel, Céline Hudelot, Mattis Hartwig
@conference{2023-45,
title: "How to Think About Benchmarking Neurosymbolic AI?",
author: "Johanna Ott and Arthur Ledaguenel and Céline Hudelot and Mattis Hartwig",
year: "2023"
booktitle: "17th International Workshop on Neural-Symbolic Learning and Reasoning -NESY 2023"
pages: "1 -- 8"
}
Improving Patient Trajectory Forecasts in Hospitals: Using Emergency Department Data for Length of Stay Prediction and Next Hospital Unit Classification
Autor:innen: Alexander Winter, Toralf Kirsten, Mattis Hartwig
Improved Techniques for Training Tabular GANs Using Cramer’s V Statistics
Autor:innen: Melle Mendikowski, Benjamin Schindler, Thomas Schmid, Ralf Möller, and Mattis Hartwig
@conference{Mendikowski2023-7,
title: "Improved Techniques for Training Tabular GANs Using Cramer’s V Statistics",
author: "Melle Mendikowski and Benjamin Schindler and Thomas Schmid and Ralf Möller and and Mattis Hartwig",
year: "2023"
booktitle: "Proceedings of the Canadian Conference on Artificial Intelligence"
pages: "1 -- 12"
}
Evaluating the Effects of a Priori Deep Learning Image Synthesis on Multi-Modal MR-to-CT Image Registration Performance
Autor:innen: Nils Frohwitter, Alessa Hering, Ralf Möller, Mattis Hartwig
@conference{Frohwitter2023-17,
title: "Evaluating the Effects of a Priori Deep Learning Image Synthesis on Multi-Modal MR-to-CT Image Registration Performance",
author: " Nils Frohwitter and Alessa Hering and Ralf Möller and Mattis Hartwig",
year: "2023"
booktitle: "Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2023)"
pages: " 322 -- 329"
}
Predicting Hospital Length of Stay of Patients Leaving the Emergency Department
Autor:innen: Alex Winter, Toralf Kirsten, Mattis Hartwig
@conference{Winter2023-25,
title: "Predicting Hospital Length of Stay of Patients Leaving the Emergency Department",
author: "Alex Winter and Toralf Kirsten and Mattis Hartwig",
year: "2023"
booktitle: "Proceedings of the 16th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2023) - HEALTHINF"
pages: "124 -- 131"
}
New Methods for Efficient Query Answering in Gaussian Probabilistic Graphical Models
Autor:innen: Mattis Hartwig
@phdthesis{Hartwig2022-40,
title: "New Methods for Efficient Query Answering in Gaussian Probabilistic Graphical Models",
author: "Mattis Hartwig",
year: "2022"
}
Creating Customers That Never Existed: Synthesis of E-commerce Data Using CTGAN
Autor:innen: Melle Mendikowski, Mattis Hartwig
@conference{Mendikowski2022-30,
title: "Creating Customers That Never Existed: Synthesis of E-commerce Data Using CTGAN",
author: "Melle Mendikowski and Mattis Hartwig",
year: "2022"
booktitle: "18th International Conference on Machine Learning and Data Mining MLDM"
pages: "91 -- 105"
}
Increasing State Estimation Accuracy in the Inference Algorithm on a Hybrid Factor Graph Model
Autor:innen: Mattis Hartwig, Tanya Braun, Ralf Möller
@conference{Hartwig2022-5,
title: "Increasing State Estimation Accuracy in the Inference Algorithm on a Hybrid Factor Graph Model",
author: "Mattis Hartwig and Tanya Braun and Ralf Möller",
year: "2022"
booktitle: "The International FLAIRS Conference Proceedings"
pages: ""
}
Lifted Division for Lifted Hugin Belief Propagation
PDF ansehenAutor:innen: Moritz P. Hoffmann, Tanya Braun, Ralf Möller
@conference{Hoffmann2022-42,
title: "Lifted Division for Lifted Hugin Belief Propagation",
author: "Moritz P. Hoffmann and Tanya Braun and Ralf Möller",
year: "2022"
booktitle: ""
pages: "6501 -- 6510"
}
Handling Overlaps When Lifting Gaussian Bayesian Networks
Autor:innen: Mattis Hartwig, Tanya Braun, Ralf Möller
@conference{Hartwig2021-29,
title: "Handling Overlaps When Lifting Gaussian Bayesian Networks",
author: "Mattis Hartwig and Tanya Braun and Ralf Möller",
year: "2021"
booktitle: "Thirtieth International Joint Conference on Artificial Intelligence"
pages: ""
}
Cooperation and Social Rules Emerging From the Principle of Surprise Minimization
Autor:innen: Mattis Hartwig, Achim Peters
@article{Hartwig2021-13,
title: "Cooperation and Social Rules Emerging From the Principle of Surprise Minimization",
author: "Mattis Hartwig and Achim Peters",
year: "2021"
journal: "Frontiers in Psychol."
pages: ""
}
Lifted Query Answering in Gaussian Bayesian Networks
Autor:innen: Mattis Hartwig, Ralf Möller
@conference{Hartwig2020-25,
title: "Lifted Query Answering in Gaussian Bayesian Networks",
author: "Mattis Hartwig and Ralf Möller",
year: "2020"
booktitle: "International Conference on Probabilistic Graphical Models"
pages: "233 -- 244"
}
Constructing Gaussian Processes for Probabilistic Graphical Models
PDF ansehenAutor:innen: Mattis Hartwig, Marisa Mohr, Ralf Möller
@conference{Hartwig2020-3,
title: "Constructing Gaussian Processes for Probabilistic Graphical Models",
author: "Mattis Hartwig and Marisa Mohr and Ralf Möller",
year: "2020"
booktitle: "FLAIRS Conference"
pages: "57 -- 62"
}
How to Encode Dynamic Gaussian Bayesian Networks as Gaussian Processes?
Autor:innen: Mattis Hartwig, Rolf Möller
@conference{Hartwig2020-11,
title: "How to Encode Dynamic Gaussian Bayesian Networks as Gaussian Processes?",
author: "Mattis Hartwig and Rolf Möller",
year: "2020"
booktitle: "Advances in Artificial Intelligence"
pages: ""
}
Efficient Query Answering in Nonparametric Probabilistic Graphical Models
Autor:innen: Mattis Hartwig
@conference{Hartwig2020-8,
title: "Efficient Query Answering in Nonparametric Probabilistic Graphical Models",
author: "Mattis Hartwig",
year: "2020"
booktitle: "German Conference on Artificial Intelligence"
pages: "1 -- 5"
}
New Approaches in Ordinal Pattern Representations for Multivariate Time Series
PDF ansehenAutor:innen: Marisa Mohr, Florian Wilhelm, Mattis Hartwig, Ralf Möller, Karsten Keller
@conference{Mohr2020-41,
title: "New Approaches in Ordinal Pattern Representations for Multivariate Time Series",
author: "Marisa Mohr and Florian Wilhelm and Mattis Hartwig and Ralf Möller and Karsten Keller",
year: "2020"
booktitle: "International Florida Artificial Intelligence Research Society Conference"
pages: "124 -- 129"
}
Approximate Query Answering in Complex Gaussian Mixture Models
Autor:innen: Mattis Hartwig, Marcel Gehrke, Ralf Möller
@conference{Hartwig2019-31,
title: "Approximate Query Answering in Complex Gaussian Mixture Models",
author: "Mattis Hartwig and Marcel Gehrke and Ralf Möller",
year: "2019"
booktitle: "IEEE International Conference on Big Knowledge"
pages: "81 -- 86"
}