Roman Hornungمشاهده پروفایل
پژوهشگر ارشد
- Random Forests
- Statistical Methodology for Cross-Lab Prediction
- Statistical Methodology for Multi-Omics Data
- +۵ مورد دیگر
Dr. Roman Hornung is a Research Fellow at the Institute of Medical Informatics, Biometry and Epidemiology (IBE) within Ludwig-Maximilians-Universität Munich (LMU), specializing in Biometry in Molecular Medicine. His work focuses on advancing statistical and machine learning methodologies for biomedical applications. Academic Rank: Research Fellow Key Research Areas: Random Forests, Multi-Omics Data Integration, Interpretable Machine Learning, Cross-Validation Techniques, Biomedical Informatics, and Meta-Research on Prognostic Modeling Teaching: Leads seminars on applied statistics in epidemiology and contributes to clinical epidemiology tutorials Software Contributions: Maintainer of R packages like diversityForest and prioritylasso His research bridges statistical theory and practical biomedical applications, with a particular emphasis on reproducibility and robustness in predictive modeling. He has developed novel random forest variants for multi-class outcomes and multi-omics data. Hornung’s recent publications and submitted work highlight trends in benchmarking machine learning methods, improving cross-study prediction, and understanding the impact of data preparation on model performance. Notably, he has contributed to methodologies for generalization error estimation, variable importance measures, and diversity-enhanced forest algorithms. As a member of the Biometry in Molecular Medicine working group, he collaborates on interdisciplinary projects integrating clinical and omics data. His habilitation at LMU Munich (2024) and editorial roles at Biometrical Journal and Scientific Reports underscore his academic leadership in statistical methodology for health sciences.








