Data analytics
Disturber shows building 42

Data analytics

Prof. Dr. habil. Rouven E. Haschka

Dr. Samyajoy Pal

Research Associate

Dr. Samyajoy Pal is currently a postdoctoral researcher and research assistant at the Chair of Data Analytics in the Faculty of Business Studies and Economics at RPTU Kaiserslautern-Landau.

He holds a PhD in Statistics from the Department of Statistics at LMU Munich. His research is at the intersection of statistics, machine learning and artificial intelligence with a particular focus on the methodological and theoretical foundations of statistical learning, mathematical statistics and statistical inference.

His research focuses in particular on flexible multivariate mixed models and their applications in statistical machine learning, model-based clustering, classification and predictive modeling. Further research areas include generalized linear models, the analysis of high-dimensional and compositional data, regularized regression methods and multivariate statistical modelling.

A central goal of his research is the development of interpretable and probabilistically based machine learning methods that combine classical statistical inference with modern AI approaches. In this context, he deals with Explainable Artificial Intelligence (XAI), interpretable machine learning as well as deep neural networks based on Variational Autoencoders (VAEs) and mixture-based latent variable models. His research aims to strengthen the statistical foundations of modern AI through theoretically sound, interpretable and inference-based learning methods.

Research interests

Statistical Learning, Statistical Inference, Mathematical Statistics, Interpretable Machine Learning, Explainable Artificial Intelligence, Multivariate Mixed Models, Model-based Clustering, High-dimensional Data Analysis

Further information

  • Pal, S. (2025). Advances in finite mixture models with applications to unsupervised learning. PhD Dissertation, LMU Munich.

  • Pal, S., & Heumann, C. (2025). Revisiting Dirichlet Mixture Model: unraveling deeper insights and practical applications. Statistical Papers, 66 (1), 2.

  • Pal, S., & Heumann, C. (2024). Flexible Multivariate Mixture Models: A Comprehensive Approach for Modeling Mixtures of Non-Identical Distributions. International Statistical Review.

  • Pal, S., & Heumann, C. (2024). Gene coexpression analysis with Dirichlet mixture model: accelerating model evaluation through closed-form KL divergence approximation using variational techniques. In: International Workshop on Statistical ModelingSpringer Nature Switzerland, 134-141.

  • Pal, S., & Heumann, C. (2024). Gaussian mixture model with modified hard EM algorithm in clustering problems. In: Statistical Modeling and Applications on Real-Time ProblemsCRC Press, 153-179.

  • Pal, S., & Heumann, C. (2022). Clustering compositional data using Dirichlet mixture model. PLOS One, 17 (5), e0268438.