Vortrag im Rahmen des University of Vienna Physics Colloquiums
The use of artificial intelligence (AI) in research is changing scientific practices and is likely to transform science in the future. For basic science it is desirable if not mandatory that its methods remain transparent and that humans are able understand how a problem was solved. I believe that the physical sciences, with their specific standards for explanation, can serve as a reference for the development of interpretable and explainable AI. In this talk, I will review some of our research on the use and development of AI in quantum physics and beyond. This includes the model of projective simulation (PS), which is an interpretable model for learning and decision-making which has applications, e.g., in the design of quantum experiments, robotic skill acquisition and the modelling of biological behaviour. I will review some of these applications and discuss the interplay of concepts from quantum information and machine learning, more generally. These studies also address the problem of artificial agency in quantum physics.
A light lunch buffet will be offered before the lecture at around 13:00.
