The Rhineland-Palatinate University of Technology (RPTU) has launched the “Machine Learning Kaiserslautern-Landau” (MLKL) initiative to bundle and strengthen its research efforts in the field of machine learning. The main goal is to achieve an international leadership position in machine learning for sustainable production and the environment. To this end, new methods are being developed in the areas of unsupervised deep learning, learning on complex structured data types and explainable AI.

Particular attention is paid to anomaly detection and the development of generative AI models for complex data structures such as time series and process data, which are often heterogeneous and have different formats and scales. In addition, work is being done on transparent and interpretable models to increase confidence in predictions and make decision-making processes comprehensible.

These methods can be used in areas such as the monitoring and optimization of energy and resource consumption in factories and the prediction of environmental impacts. By analyzing data sources such as satellite images, biological monitoring data and chemical applications, more precise predictions can be made and environmental impacts better understood.

MLKL thus not only strengthens the research profile of RPTU, but also supports state policy priorities, particularly in the areas of environment and economy. The integration of explainable AI into MLKL’s research agenda also contributes to its social relevance.