The Young Investigator Network is the platform and democratic representation of interests for independent junior research group leaders and junior professors at the Karlsruhe Institut of Technology.


Visit the News Archive 2022 to learn what YIN members have recently achieved.

ERC StG 2022
Heinz Maier-Leibniz Prize 2022
Harald Perten Prize 2022
Gaede Award 2022
ERC StG for Julian QuintingAmadeus Bramsiepe, KIT
ERC Starting Grant for Making Weather Forecasts more Precise and Efficient

With the project ASPIRE, Julian Quinting aims at Advancing Subseasonal PredIctions at Reduced computational Effort. The idea is to make more use of sources in the atmospheric system with high intrinsic predictability such as recurring patterns of tropical convection that vary on the time scale of two weeks to two months. Patterns of convection in the Pacific, for example, have a major influence on the weather in Europe but are insufficiently represented in weather prediction models. To improve this, while keeping computing efforts at a minimum, Quinting develops machine learning models to imitate the effects of high resolution. The European Research Council will fund the project for five years.

KIT press release
Pitfalls in securityArp, TU Berlin
Pitfalls in the use of artificial intelligence in computer security

Machine learning (ML) is often superior to traditional methods. When used in computer security, however, it is prone to subtle pitfalls, as Christian Wressnegger 's team has discovered together with international partners. "For example, a learning antivirus program trained on incomplete data could prove useless in practice," Wressnegger explains. The researchers examined 30 recent papers that used ML for IT security and were published at prestigious computer and system security conferences. All had failed to address one or more sources of error. "There is a lack of awareness of the difficulties of applying machine learning correctly," says the expert for cyber security.

Dos and Don’ts of ML in Computer Security
KI-basierte Methodik für die schnelle Ertüchtigung unreifer Produktionsprozessewbk/KIT
DFG Research Group: Making Production Processes Faster Usable with AI

To bring new products faster to market, companies need to improve immature production processes on the fly. The goal of the new DFG research group at KIT is to make process adjustments cheaper, faster and more efficient with the systematic use of artificial Intelligence (AI). In the subproject "Management and Quantification of Process Maturity Improvement", Tobias Käfer uses knowledge graph-based methods to model data and knowledge around the process and make it available to experts. To address challenges like the distribution of data, its heterogeneity, and the treatment of physical knowledge, the scientists combine semantic data processing with qualitative reasoning.