DFG Priority Program on Machine Learning

April 01, 2021
Prof. Dr. Karsten Reuter, director of the Department of Theory, is co-coordinator of a new priority program funded by the German Research Foundation (DFG). The program, based at Westfälische Wilhelms-Universität (WWU) Münster, investigates the use and development of machine learning for molecular applications.

The German Research Foundation (DFG) established 13 new priority programs (SPPs) for 2022. The 13 new collaborations, which were selected from 47 submitted initiatives, will receive a total of around 82 million euros for an initial three years. The priority programs are intended to investigate the scientific foundations of current or emerging research areas. All programs have a strong interdisciplinary focus and are characterized by the use of innovative methods.

This is also the case with the priority program on Machine Learning. Prof. Dr. Karsten Reuter, Director of the Theory Department at the Fritz Haber Institute, is one of the co-coordinators of the program "Use and Development of Machine Learning for Molecular Applications - Molecular Machine Learning", along with Prof. Dr. Jürgen Bajorath of the B-IT / LIMES Institute at the University of Bonn. The project is led by chemist Prof. Dr. Frank Glorius from the Organic Chemistry Institute of the Westphalian Wilhelms University (WWU) Münster. They represent three core areas of the interdisciplinary collaborative research program - molecular and drug development, experimental and theoretical chemistry.

The focus of the project is on molecular problems such as the prediction of chemical reactions or the development of new algorithms for modeling molecular properties. The aim is to develop tools which, on the one hand, help to understand molecular relationships (ExAI - "explainable artificial intelligence") and, on the other hand, model molecular behavior in such a way that they support laboratory chemists in their everyday work. The long-term goal is to use artificial intelligence to process simple tasks automatically and comprehensibly, thereby accelerating the development of analytical methods, new reactions or drugs.

A core goal of this program is collaboration and networking. "We started bringing together the scientists involved in this back in 2020 and got a lot of positive feedback. Now we want to use the program to give this important future topic a push forward," says Frank Glorius. “Particularly important is to bring together experimentalists and theoreticians, and to increasingly apply machine learning to experimental data”, adds Karsten Reuter.

Priority programs are funded for six years. The DFG invites interested scientists to submit proposals to participate in all SPPs.

Other Interesting Articles

Go to Editor View