Center for Trustworthy Data Science and Security
Contact (Office)
| N.N. |
|
rctrust.kleinlab.fk05@tu-dortmund.de
|
|
0231 7557820 |
|
Joseph-von-Fraunhofer-Str. 25 44227 Dortmund Germany |
Welcome to the Chair of Uncertainty Quantification and Statistical Learning at the Research Center Trustworthy Data Science and Security (UA Ruhr) and the Department of Statistics (Technische Universität Dortmund). In addition to teaching graduate courses on Bayesian statistics and machine learning, we are involved in several research projects.
Feel free to follow us on Twitter!
Information for Students
- The office hours of Prof. Klein are by appointment.
- For details regarding available Bachelor's and Master's Theses, please see the corresponding teaching section.
News
- PhD student Paul Bach presented a poster about "Spatially adaptive Bayesian P-splines" at IWSM 2023 (joint work with Nadja Klein). The corresponding short paper will appear shortly in the conference proceedings. (24.07.2023)
- The arXiv preprint of the new review "Distributional Regression for Data Analysis" by Nadja Klein can be found here. (21.07.2023)
- The paper “Accounting for time dependency in meta-analyses of concordance probability estimates” by Matthias Schmid, Tim Friede, Nadja Klein and Leonie Weinhold was accepted for publication in Research Synthesis Methods. (11.07.2023)
- The arXiv preprint of the new paper "Heat increases experienced racial segregation in the United States" by Till Baldenius, Nicolas Koch, Hannah Klauber and Nadja Klein can be found here. (27.06.2023)
- The paper “Informed Priors for Knowledge Integration in Trajectory Prediction” by Christian Schlauch, Nadja Klein and Christian Wirth has now been accepted in the research track of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD) 2023, congratulations! (07.06.2023)
- The arXiv preprint of the new paper "Truly Multivariate Structured Additive Distributional Regression" by Lucas Kock and Nadja Klein can be found here. (06.06.2023)
- The arXiv preprint of the new paper "The Deep Promotion Time Cure Model" by Victor Medina-Olivares, Stefan Lessmann and Nadja Klein can be found here. (24.05.2023)
- Our lead Prof. Nadja Klein reveals what she appreciates most about her field in the book "Young Scientists" by Die Junge Akademie. You can check out all featured researchers here.
- The arXiv preprint of the new paper "Dropout Regularization in Extended Generalized Linear Models based on Double Exponential Families" by Benedikt Lütke Schwienhorst, Lucas Kock, David J. Nott and Nadja Klein can be found here. (13.05.2023)
- The paper “Marginally calibrated response distributions for end-to-end learning in autonomous driving” by Clara Hoffmann and Nadja Klein was published in The Annals of Applied Statistics, congratulations! (02.05.2023)
- Our lead Prof. Nadja Klein has started her professorship at the Research Center Trustworthy Data Science and Security and TU Dortmund. Our lab is bridging statistics, maths, and computer science to offer new perspectives on statistical learning. You can find the official press release with more information (in German) here. (29.04.2023)
- We are co-hosting the AI Keynote Series at LMU Munich. You can register now and join every Thursday to gain insights from renowned guest speakers about current trends in AI research. (29.04.2023)
- Our lead Prof. Nadja Klein is now an associated member of the Integrative Research Institute on Transformations of Human-Environment Systems (IRI THESys) which conducts inter- and transdisciplinary research on paths towards sustainable futures. (29.04.2023)
- The new children's book Young Scientists by Die Junge Akademie showcases successful scientists' career paths - including our group lead Prof. Nadja Klein. It will be available from May 15, 2023. (25.04.2023)
- We welcome Maarten Jung as research assistant and Annalena Weissert as student assistant to our group. (01.04.2023)
- The paper "Approximate Bayesian computation for parameter identification in computational mechanics" by Matthias G. R. Faes, Nadja Klein, Markus Pauly, Marcos A. Valdebenito and Mauricio A. Misraji was accepted to the 14th International Conference on Applications of Statistics and Probability in Civil Engineering (ICASP14). (28.03.2023)
- We have job openings for a PhD position and PostDoc position in our group! (27.02.2023)
- The paper "Boosting Multivariate Structured Additive Distributional Regression Models" by Annika Strömer, Nadja Klein, Christian Staerk, Hannah Klinkhammer and Andreas Mayr was accepted for publication in Statistics in Medicine. The arXiv preprint can be found here. (24.02.2023)
- Clara Hoffmann is going to present at the PyCon DE & PyData Berlin 2023 (17. April - 19. April 2023) on debugging custom PyTorch models in a structured manner. More details regarding her talk can be found here. (21.02.2023)
- The paper "Bayesian Conditional Transformation Models" by Manuel Carlan, Thomas Kneib and Nadja Klein was accepted for publication in the Journal of the American Statistical Association. The arXiv preprint can be found here. (21.02.2023)
- The paper "Deep Distributional Time Series Models and the Probabilistic Forecasting of Intraday Electricity Prices" by Nadja Klein, Michael Stanley Smith and David J. Nott (Journal of Applied Econometrics) can be found here. (21.02.2023)
- The paper "Modeling Intra-Annual Tree Stem Growth with a Distributional Regression Approach for Gaussian Process Responses" by Hannes Riebl, Nadja Klein and Thomas Kneib was accepted at Journal of the Royal Statistical Society, Series C (Applied Statistics). (21.01.2023)
- A new working paper "Scalable Estimation for Structured Additive Distributional Regression" by Nikolaus Umlauf, Johannes Seiler, Matthias Wetscher, Thorsten Simon, Stefan Lang and Nadja Klein can be found here. (13.01.2023)
- The paper "Semi-Structured Distributional Regression" by David Ruegamer, Chris Kolb and Nadja Klein was accepted at The American Statistician. (01.01.2023)
- We welcome Ivan Ustyuzhaninov as postdoctoral researcher, Clara Hoffmann as PhD student and Tim-Moritz Bündert as research assistant to our group. (01.01.2023)
- We will move from Humboldt-Universität zu Berlin to TU Dortmund in April 2023! (01.01.2023)
- New working paper "Accounting for Time Dependency in Meta-Analyses of Concordance Probability Estimates" by Matthias Schmid, Tim Friede, Nadja Klein and Leonie Weinhold can be found here. (03.12.2022)
- New working paper "Anisotropic multidimensional smoothing using Bayesian tensor product P-splines" by Paul Bach and Nadja Klein can be found here. (29.11.2022)
- New working paper "Informed Priors for Knowledge Integration in Trajectory Prediction" by Christian Schlauch, Nadja Klein and Christian Wirth can be found here. (01.11.2022)
- New working paper "Distributional Adaptive Soft Regression Trees" by Nikolaus Umlauf and Nadja Klein can be found here. (19.10.2022)
Team
| Prof. Dr. Nadja Klein |  |
Group Lead | ___________ |
| Paul Bach |  |
PhD Student | |
| Guillermo Briseno-Sanchez |  |
PhD Student | |
| Tim-Moritz Bündert |  |
Research Assistant | |
| Ekin Celikkan |  |
PhD Student | |
| Clara Hoffmann |  |
PhD Student | |
| Maarten Jung |  |
Research Assistant | |
| Lucas Kock |  |
PhD Student | |
| Victor Medina-Olivares, PhD |  |
Postdoctoral Researcher | |
| Pallavi Mitra |  |
PhD Student | |
| Christian Schlauch |  |
PhD Student | |
| Bettina Schmidt |  |
PhD Student | |
| Michael Stanley Smith |  |
Mercator Fellow | |
| Ivan Ustyuzhaninov |  |
Postdoctoral Researcher | |
| Annalena Weissert |  |
Student Assistant |
Research
Research Interests
The group's research interests broadly lie at the intersection of machine learning and traditional statistical methods. This includes Bayesian Computational Methods, Bayesian Deep Learning, Machine Learning, Smoothing, Regularization and Shrinkage, Distributional Regression, Network Analysis as well as Spatial Statistics.
Ongoing Third-Party Research Projects
Prof. Klein is principal investigator in the following research projects.
- from 2023: DFG Research Unit Fusing Deep Learning and Statistics towards Understanding Structured Biomedical Data (sub-project P5, sub-project P6)
- since 2021: Experiment (funded by the Volkswagenstiftung)
- since 2020: DFG individual grant (jointly with Andreas Mayr, University of Bonn)
- since 2019: Emmy Noether Research Group (funded by the DFG)
Publications
See the Google Scholar page of Prof. Klein for a complete list. Below you can find some key publications and recent working papers.
Selected Publications
-
Bayesian conditional transformation models pdf
M. Carlan, T. Kneib and N. Klein
To appear in Journal of the American Statistical Association, 2023 -
Marginally calibrated response distributions for end-to-end learning in autonomous driving pdf
C. Hoffmann and N. Klein
Annals of Applied Statistics, 17(2), 1740-1763, 2023 -
Multivariate Conditional Transformation Models pdf
N. Klein, T. Hothorn, L. Barbanti and T. Kneib
Scandinavian Journal of Statistics, 49(1), 116-142, 2022 -
Marginally Calibrated Deep Distributional Regression pdf
N. Klein, D. J. Nott and M. S. Smith
Journal of Computational and Graphical Statistics, 30(2), 467-483, 2021 -
Bayesian Variable Selection for Non-Gaussian Responses: A marginally-calibrated Copula Approach pdf
N. Klein and M. S. Smith
Biometrics, 77(3), 809-823, 2021 -
Bayesian Inference for Regression Copulas pdf
M. S. Smith and N. Klein
Journal of Business and Economic Statistics, 39(3), 712-728, 2021 -
Modelling Regional Patterns of Inefficiency: A Bayesian Approach to Geoadditive Panel Stochastic Frontier Analysis with an Application to Cereal Production in England and Wales pdf
N. Klein, H. Herwartz and T. Kneib
Journal of Econometrics, 214(2), 513-539, 2020 -
Implicit Copulas from Bayesian Regularized Regression Smoothers pdf
N. Klein and M. S. Smith
Bayesian Analysis, 14(4), 1143-1171, 2019 -
BAMLSS: Bayesian Additive Models for Location, Scale and Shape (and Beyond) pdf
N. Umlauf, N. Klein and A. Zeileis
Journal of Computational and Graphical Statistics, 27(3), 612-627, 2018 -
Bayesian Generalized Additive Models for Location, Scale and Shape for Zero-Inflated and Overdispersed Count Data pdf
N. Klein, T. Kneib and S. Lang
Journal of the American Statistical Association, 110(509), 405-419, 2015 -
Bayesian Structured Additive Distributional Regression with an Application to Regional Income Inequality in Germany pdf
N. Klein, T. Kneib, S. Lang and A. Sohn
Annals of Applied Statistics, 9(2), 1024-1052, 2015 -
Bayesian Conditional Transformation Models pdf
M. Carlan, T. Kneib and N. Klein
To appear in Journal of the American Statistical Association. -
Deep Distributional Time Series Models and the Probabilistic Forecasting of Intraday Electricity Prices pdf
N. Klein, M. S. Smith and D. J. Nott
To appear in Journal of Applied Econometrics. -
Learning Causal Graphs in Manufacturing Domains Using Structural Equation Model
M. Kertel, S. Harmeling, M. Pauly and N. Klein
To appear in International Journal of Semantic Computing. -
Estimating Uncertainty in Real Estate Valuation: A Multilevel Approach based on Distributional and Quantile Regression
A. Razen, W. Brunauer, N. Klein, T. Kneib, S. Lang and N. Umlauf
To appear in Statistical Modelling. -
Boosting Multivariate Structured Additive Distributional Regression Models pdf
A. Strömer, N. Klein, C. Staerk, H. Klinkhammer and A. Mayr
To appear in Statistics in Medicine. -
Semi-Structured Distributional Regression pdf
D. Rügamer, C. Kolb and N. Klein
To appear in The American Statistician. -
Gaussian Process Responses in Distributional Regression: A versatile Model Class with an Application to Tree Growth Dynamics
H. Riebl, N. Klein and T. Kneib
To appear in Journal of the Royal Statistical Society, Series C.
Working Papers
-
Dropout regularization in extended generalized linear models based on double exponential families pdf
B. Lütke Schwienhorst, L. Kock, D. Nott and N. Klein, 2023
-
The deep promotion time cure model pdf
V. Medina-Olivares, S. Lessmann and N. Klein, 2023
-
Bannmi deciphers potential n-to-1 information transduction in signalling pathways to unravel message of intrinsic apoptosis pdf
B. Schmidt, C. Sers and N. Klein, 2023
Submitted. Available on request. -
Scalable estimation for structured additive distributional regression pdf
N. Umlauf, J. Seiler, M. Wetscher, T. Simon, S. Lang and N. Klein, 2023
-
Farm Structure and Environmental Context drive Farmers' Decisions on the Allocation of Ecological Focus Areas in Germany
V. Alarcón-Segura, S. Roilo, A. Paulus, M. Beckmann, N. Klein and A. Cord, 2022
Submitted. Available on request. -
Anisotropic Multidimensional Smoothing using Bayesian Tensor Product P-Splines pdf
P. Bach and N. Klein, 2022
-
MultiFlags: Multi-Class Size Recommendations for Fashion E-Commerce
H. Boeddeker, N. Klein, L. Lefakis, A. Weffer, L. Karessli and M. Shirvany, 2022
Submitted. Available on request. -
Incorporating Actor Heterogeneity Into Large Network Models through Variational Approximations pdf
N. Klein and G. Kauermann, 2022
-
Intergenerational Social Mobility in the United States: A Multivariate Analysis using Bayesian Distributional Regression
A. März, N. Klein, T. Kneib and O. Mußhoff, 2022
Submitted. Available on request. -
A Multidimensional Approach is needed to better quantify Land-Use Intensity in Biodiversity Models
S. Roilo, A. Paulus, V. Alarcón Segura, L. Kock, M. Beckmann, N. Klein and A. Cord, 2022
Submitted. Available on request. -
Informed Priors for Knowledge Integration in Trajectory Prediction pdf
C. Schlauch, N. Klein and C. Wirth, 2022
Submitted. -
Accounting for Time Dependency in Meta-Analyses of Concordance Probability Estimates pdf
M. Schmid, L. Weinhold, N. Klein and T. Fried, 2022
-
Distributional Adaptive Soft Regression Trees pdf
N. Umlauf and N. Klein, 2022
Submitted. -
Scalable Estimation for Structured Additive Distributional Regression pdf
N. Umlauf, J. Seiler, M. Wetscher, T. Simon, S. Lang and N. Klein, 2022
Submitted. -
Posterior Concentration Rates for Bayesian O'Sullivan Penalized Splines pdf
P. Bach and N. Klein, 2021
-
Bayesian Effect Selection for Additive Quantile Regression with an Analysis to Air Pollution Thresholds pdf
N. Klein and J. Mateu, 2021
R& R Annals of Applied Statistics -
Flexible Specification Testing in Semi-Parametric Quantile Regression Models pdf
T. Kutzker, N. Klein and D. Wied, 2021
-
Bivariate Analysis of Birth Weight and Gestational Age depending on Environmental Exposures: Bayesian Distributional Regression with Copulas pdf
J. Rathjens, A. Kolbe, J. Hölzer, K. Ickstadt and N. Klein, 2021
-
Approximate Bayesian Computation for Parameter Identification in Computational Mechanics
M. G. R. Faes, N. Klein, M. Pauly, M. A. Valdebenito and M. A. Misraji, 2023
Submitted to the 14th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP14.
Teaching
-
Master Theses
We offer Master theses at the intersection of statistics and machine learning. Please see below for titles of available theses in these areas. If you are interested in one of these projects and fulfill the requirements laid out in our thesis guidelines, please contact us.
- Understanding And And Implementing Predictive Information Criteria For Bayesian Models
- Asymptotic Behaviour Of The Posterior In Overfitted (Deep) Mixture Models
- Posterior Concentration Rates For Bayesian High-Dimensional Sparse Additive Models
- Using Posterior Predictive Checks With Semistructured Deep Distributional Regression
- Efficient Estimation Of Distributional Regression Models Using Hmc Based On Normalizing Flows
- Predicting Revenue Streams Of Small And Medium-Sized Enterprises
- Standard Errors Of Variational Inference
- Uncertainty Quantification For Deep Learning
- Deep Gaussian Process Modelling
- Using Stacking To Average Distributional Regression Models
- Bayesian Nonparametric Conditional Density Estimators
- Distributional Joint Modelling
- Approximations Of Normalizing Constants In Doubly-Intractable Likelihoods
- Measuring The Explained Variance In Structured Additive Distributional Regression
- Comparisons And Implementation Of Non-Local Shrinkage Priors
- A Scalable Joint Model Of Longitudinal And Survival Data With A Time-Varying Association
- A Unified Deep-Cure Model Approach
- Admissible Learning Algorithms
- Fast Expectation Maximization Algorithm For Spatial Promotion Time Cure Models
- Approximated Bayesian Multivariate Spatial Factor Analysis
- A Scalable Joint Model Of Longitudinal And Survival Data With Long-Term Survivors
- Leverage Generative Models To Improve Performance In Unbalanced Tabular Settings
- Exact Likelihood Inference In Joint Modelling Settings
- Targeted Sampling Based On Epistemic Uncertainty For Improving Predictive Performance Of DL Tasks In The Case Of Costly Data Acquisition
- Uncertainty Quantification For Complex-Valued Deep Neural Networks
- Improved Variational Approximations
- Non-Gaussian Regression Copulas
- Variational Inference for Distributional Regression
Theses with External Collaborators
- Analysis of routine data from outpatient medical care (in cooperation with the University of Lübeck)
A selection of completed theses at the chair can be found below.
- Analyzing Heat and Experienced Racial Segregation using Large-Scale Foot Traffic Data
- A Python Implementation for the Structural Topic Model
- Optimierung der Kraftwerkssteuerung mittel Reinforcement Learning unter Einsatz von kurzfristigen Ausgleichsenergiepreiseprognosen
- Including Deep Neural Network Architectures into Multistage Intensity Models: An Application to Credit Risks
- MultiFlags and LatentFlags: A Probabilistic Framework for Size Advice in Fashion E-Commerce
- Interpretable Modelling of ICU Patients Remaining Length-of-Stay Distribution using Tabular Patient Data, Clinical Notes and Irregularly Spaced Clinical Measurements
- Using Variational Inference to Estimate Structred Additive Distributional Regression Models
- Elastic Full Procluster Means for Sparse and Irregular Curves
- Reconstructing Multivariate Functional Data with Medical Applications
- Investment Constraints in Southern Europe: A Spatial Econometric Analysis of World Bank Enterprise Surveys
- Application of Regression Trees on Compositional Data Using European Parliament Election Results
- Die Modellierung der COVID-19 Fallzahlen in Abhängigkeit von Strukturdaten zu Wetter und Bevölkerung in Berlin
- Vergleich von Vorhersagemodellen zu Stornierungen von Hotelbuchungen
Master Theses
Bachelor Theses
-
Lectures (Summer Semester 2023)
Advanced Bayesian Data Analysis
Lecture: Thursdays 10:00 - 12:00 (Mathetower E25)
Exercise: Thursdays 12:00 - 14:00 (Mathetower E25)
More details in first session (06.04.2023)