Wolfram Luther

XAI Research Directions and Challenges within the Public Sector and Recent 4.0 Initiatives: An Overview


Wolfram Luther, is a senior professor in the Department of Computer Graphics, Image Processing, and Scientific Computing at the University of Duisburg-Essen in Duisburg. Luther received his doctorate in mathematics from the University of Marburg in 1973, after studying in Marburg and Freiburg from 1966 to 1970. He habilitated at RWTH Aachen University in 1980. After several academic activities and guest professorships in Aachen, Mulhouse, Ulm, and Duisburg, he became a professor of computer science in 1993 at the Gesamthochschule Duisburg, which merged into the University of Duisburg-Essen in 2003. Luther has chaired the examination board for the diploma program in Applied Computer Science since its creation in 2000. He has also worked for the German National Academic Foundation at the UDE. In 2004 and 2010, he held visiting professorships at the Universidad de Chile.

Description of the Talk:

The engagement with explainable artificial intelligence research directions and challenges stems from a suggestion by Ashot Harutyunyan and a discussion on this topic in the spring as part of the Fast project.This talk will describe what was found out, first, but not a final answer to the question of which models, methods and application areas, requirements, and optimization goals fit together in smart environments and which don’t, which intelligent sensors and interfaces are needed for control, risk avoidance, human understandable processing and result communication, and sometimes how performance, efficiency, and accuracy under uncertainty can be evaluated using formalized rigorous evaluation metrics.