The first AI developer working group of the year took place this Monday. Prof. Dr. Emmanuel Müller from TU Dortmund University visited us with a lecture on “Trustworthy Machine Learning: Explainable & Verifiable Anomaly Detection”.
The core topic was the processes by which algorithms can perform reliable anomaly detection. He identified the automotive industry, healthcare and infrastructure contexts as current areas of application for these technologies – in these areas, he considers algorithms that detect faults, anomalies and the need for repairs at an early stage to be particularly valuable. In order to increase the reliability of these algorithms, they need to be trained to not only detect anomalies, but to describe them in the best possible way. Only then will they become effective tools that can be used to fill the gap between the amount of data and the human decision-maker.
A big thank you goes to Prof. Dr. Emmanuel Müller and all those present who asked questions with interest and linked interdisciplinary approaches to the lecture. This time, the closing session turned out to be an intensive networking opportunity that many visitors took advantage of.