Admission to the Graduate Centre of the bidt
Early-career researcher Kata Vuk from the University of Regensburg uses artificial intelligence to improve learning models in the healthcare sector
September 17, 2024, press release of the UR
"Young, smart, innovative: the AI talents of today will have a decisive impact on our society of tomorrow. That's why we are supporting ten excellent postdocs and their AI projects with a total of around five million euros," announced Science Minister Markus Blume in Munich today. One of the talented researchers receiving funding is Dr Kata Vuk, a postdoc in Professor Dr Merle Behr 's research group at the Faculty of Informatics and Data Science at the University of Regensburg (UR). The mathematician's project aims to improve the application of machine learning models in the healthcare sector.
With this programme, the Bavarian State Ministry of Science, Research and the Arts provides particularly talented young AI researchers with long-term support in developing their own research profile as the basis for a long-term career in science. "Because the fact is that artificial intelligence will significantly change our lives in many areas and we want to help shape this process in line with our values," says Minister Blume.
From January 2025, the researchers, who are at the beginning of their scientific careers, will be accepted into the Graduate Centre at the Bavarian Research Institute for Digital Transformation (bidt), an institute of the Bavarian Academy of Sciences and Humanities. The funding is intended for a period of up to four years. It includes personnel funding for the postdocs, research funding and support for networking and further qualification in the field of digitalisation. The talented researchers and their proposed projects were recommended for funding by a committee of experts from outside Bavaria.
Kata Vuk's project "From Data to Discovery in the Healthcare Information Age: Interpretable Machine Learning with Piecewise Constant Models" aims to improve the application of machine learning models in the healthcare sector. "I focus on the development of models that not only enable precise predictions, but are also easy to understand. This is particularly important in the healthcare sector, as medical decisions need to be comprehensible and transparent," explains Kata Vuk, who completed her doctorate at the Chair of Probability Theory and its Applications at Ruhr-Universit?t Bochum in 2023 after studying mathematicians. She joined Merle Behr's Machine Learning working group at the University of Regensburg in June 2023. Her research findings also contribute to the Collaborative Research Centre TRR 374, which is funded by the German Research Foundation at the University of Regensburg and is researching chronic kidney disease.
The mathematician investigates piecewise constant models, in particular decision trees and change-point models. Piecewise constant models are mathematicians who represent complex relationships in a simple form. Decision trees and change-point models help to analyse data, for example to make decisions or identify changes in data patterns. Kata Vuk is developing approaches to adapt these models so that they make individualised predictions and remain comprehensible. "Decision trees, which are particularly suitable for tabular data, such as tabular patient data, offer high predictive power, but tend to lose their interpretability in complex combinations," explains Vuk.
Change-point models, which are used for serial data such as time-dependent vital signs or genetic information, are in turn optimised to ensure interpretability even at high dimensions: "This is crucial for identifying specific changes in a patient's state of health or in genetic sequences," says Vuk. "In the long term, the project aims to use such interpretable models for personalised medicine in order to develop tailored treatment strategies," adds Merle Behr, who is delighted with the success of the postdoc working at her chair.
Information/contact
The aim of the STMWK funding programme is to support particularly qualified young scientists from the first or second year after their doctorate in developing a research profile and thus qualifying them to remain in science. The Bavaria-wide Graduate Centre is coordinated by the bidt. With its diverse research projects and dialogue activities, it offers the best framework conditions for the planned projects.
Young Talents | bidt (external link, opens in a new window)
About Kata Vuk (external link, opens in a new window)
Machine Learning Group of the FIDS at the University of Regensburg
UR research field Digital Transformations (external link, opens in a new window)
UR research field Integrated Sciences in Life, Health, and Disease (external link, opens in a new window)
