Prof. Dr. Jana de Wiljes
I am the head of the research group Mathematics of Data Science at the Institute of Mathematics at TU Ilmenau. I am also a PI of the Collaborative Research Centre 1294 on Data Assimilation on projects A02, A03, B06 and B08.
I mainly am interested in the development of computational methods and the derivation of mathematical principles for sequential learning particularly in the context of Bayesian inverse problems/nonlinear filtering.
Katherine Briceno Guerrero
I am currently doing my PhD within Project B08 of SFB 1294 on Continuous learning by integrating reinforcement learning and data assimilation to individualise drug treatments. Within that project my research is mainly focused on designing new methodologies for decision support by combine sequential learning ideas with reinforcement learning tools and to derive the associate uncertainties. I am also the PhD representative of the International Research Training Group.
Marina Garcia Penaranda
I am a PhD student at GFZ (German Research Centre for Geosciences) working on the Collaborative Research Centre 1294 on Data Assimilation project B06: Novel methods for the 3D reconstruction of the dynamic evolution of the Van Allen belts using multiple satellite measurements.
I study the Earth’s Ring Current (1-100 keV), which is a is a complex dynamic system that plays an important role in geomagnetic storms. This ring-shaped current environment changes its structure and intensity on different time scales as a result from the incoming solar wind. Since particle populations display very different behaviors, is extremely hard to develop physics-based forecasting models. To tackle this issue, I am working on intercalibrating satellite data to improve the temporal and spatial resolution, which will be combined through data-assimilation techniques with physics-based models in order to get a better understanding of the underlying dynamical processes.
I’m a PhD student in the mao.docfunds doctoral school at the University Klagenfurt, Austria. My research interests are on filtering methods for high-dimensional nonlinear dynamical systems. By leveraging Bayesian inference, these methods accurately estimate system states despite complex interactions.
I have the great pleasure working together with my supervisor Jana de Wiljes and her PostDoc Gottfried Hastermann. Our objective is to conduct a comprehensive comparison of various filter methods for high-dimensional nonlinear dynamical systems. This comparison is especially crucial in a scenario where other factors, such as addressing systematic errors in modelling or ensuring data quality control, can greatly influence the assimilation system.
I have always enjoyed conversing with people, exchanging differing viewpoints, and getting fresh insights on issues and potential solutions, both in my working life and throughout my time as a student.
I am a researcher in the Collaborative Research Centre 1294 on Data Assimilation and interested in the quest of incorporating
inaccurate, and incomplete data into physics based models.
To this end, I design algorithms and provide numerical analysis in the
interplay of Bayesian non-linear filtering and structure-preserving
methods for non-linear (partial) differential equations.
The models of choice often expose phenomena on multiple scales and
originate from Hamiltonian mechanics, geophysical fluid dynamics or
space weather prediction.
Along the way, I develop research software in python and C++.
Dr. Karen Seidel
I am interested in machine and human learning of mathematics and computer science.
My current main research interest lies in statistical algorithmic learning theory, in particular the analysis of bandit algorithms with dependencies. I am also very interested in its connections to applied machine learning, cognitive science and robotics. I am jointly in the lab of Jana de Wiljes and Alexandra Carpentier.
Robotics Lab and Master Projects
I am a Data Science master’s student collaborating on research in Causal Inference for non-stationary systems. I am also interested in the general Sequential Learning paradigm, specially Reinforcement Learning.
The main goal of my research is to analyse and extend existing regime dependent causal discovery algorithms for the non-linear case. Such algorithms are relevant both in the geosciences as well as in personalised medicine, where seasons and treatments may change the underlying data generating process.
For more information on my background, check my personal website.
I am a computational science master’s student doing my Master’s Thesis on search and rescue missions in the robotics lab supervised by Dr. Jana de Wiljes. My goal is to create a probabilistic model using Bayesian inference and data assimilation to find an optimal solution. I’m looking forward to solve problems, that arise along the way, and how to tackle them.
My side project is to get more people hooked on working with robots. I hope to get a hackathon started in the near future.
I study data science as a master’s student and I work in the Robotics Lab supervised by Dr. Jana de Wiljes for the SFB 1294. Mainly my field of research is focused on Reinforcement Learning specifically for the task of search and rescue.
Pursuing this topic requires setting up the mathematical framework, implementing the algorithm as well as assembling the hardware of the robot.
We are using technologies like ROS2, SLAM and LIDAR while the implementation is done in Python, C++ and Rust.
Besides these projects we also focus on organising events to make robotics more accessible to everyone by hosting hackathons for students as well as children