Bernd Dudzik is a post-doctoral researcher in the Intelligent Systems department at the Delft University of Technology. He has obtained a Bachelor's degree in Online Media at Furtwangen University (Germany) and a Master's degree in Media Technology at Leiden University. His research is situated in the areas of Affective Computing and Social Signal Processing. It focuses on enabling technological systems to infer cognitive and affective states from data about human behavioral cues using multimodal machine learning. He is particularly interested in developing context-sensitive approaches for this kind of computer-based behavioral analysis. By accounting for person- and situation-specific influences on the underlying psychological processes driving human behavior, these have the potential for automatic inferences that are more accurate and robust across different scenarios in the real world. As part of the Hybrid Intelligence Project he is currently exploring the computational modeling of social perception processes and partner selection choices in collaborative interactions based on behavioral data about the involved individuals.