Ehsan Modiri

Mehrdad Mohannazadeh

Postdoctoral scientist

Contact


Department of Computational Hydrosystems (CHS)
Helmholtz Centre for Environmental Research - UFZ

Permoserstraße 15
04318 Leipzig

Building 7.1, Room 421

mehrdad.mohannazadeh@ufz.de

What I do


I am a Postdoc scientist in the field of hydrological modelling. My background is mainly in fields of machine Learning and applied Mathematics, therefore, I am involved in gathering and processing meteorological data that are necessary to run hydrological models for analysis and prediction purposes. Moreover, I am actively involved in modifying the models for better forecast. At UFZ, I am a part of research groups that develop the following projects: MOSES, HI-CAM II, 4DHydro, and SCENIC.

Educations


2019 - 2023 Bielefeld University, Germany
Ph.D. in Computer Sciences
Thesis: Probabilistic Methods for Robust, Reliable, and Interpretable Classification with Learning Vector Quantization
2016 - 2018 University of Applied Sciences Mittweida, Germany
M.Sc. in Applied Mathematics
2007 - 2012 Shahid Beheshti University, Iran
B.E. in Electrical Engineering (Electronics)

Publications


  • M. M. Bakhtiari, D. Staps,  and T. Villmann, “Learning vector quantization in
    context of information bottleneck theory,” in ESANN 2023, 2023
  • M. M. Bakhtiari and T. Villmann,  “The geometry of decision borders between affine space prototypes for nearest prototype classifiers,” in The 22nd International Conference on Artificial Intelligence and Soft Computing. Springer, 2023 2/3
  • M. M. Bakhtiari, A. Villmann, and T. Villmann, “An interpretable two-layered neural network structure based on component-wise reasoning,” in The 22nd International Conference on Artificial Intelligence and Soft Computing. Springer, 2023
  • T. Villmann, M. Kaden, M. Mohannazadeh Bakhtiari, and A. Villmann, “Appropriate data density models in probabilistic machine learning approaches for data analysis,” in International Conference on Artificial Intelligence and Soft Computing. Springer, 2019, pp. 443–454
  • S. Musavishavazi, M. Mohannazadeh Bakhtiari, and T. Villmann, “A mathematical model for optimum error-reject trade-off for learning of secure classification models in the presence of label noise during training,” in International Conference on Artificial Intelligence and Soft Computing. Springer, 2020, pp. 547–554
  • M. Kaden, R. Schubert, M. Mohannazadeh-Bakhtiari, L. Schwarz, and T. Villmann, “The lvq-based counter propagation network–an interpretable information bottleneck approach,” in Proceedings of the 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN’2020), Bruges (Belgium), page in this volume, Louvain-La-Neuve, Belgium, 2021, p. i6doc
  • M. M. Bakhtiari and T. Villmann, "Classification-by-component including chow’s reject option,” in International Conference on Neural Information Processing (ICONIP 2022). Springer, 2022
  • M. M. Bakhtiari and T. Villmann, “Modification of the classification-by-component predictor using dempster-shafer-theory,” in International Workshop on Self-Organizing Maps. Springer, 2022, pp. 41–52