- Diese Veranstaltung hat bereits stattgefunden.
Double Feature – Sven Mayer & Jens Schwarzbach
19. Februar, 18:30–20:30
Machine Learning Special
Machine Learning for Human-Computer Interaction
Sven Mayer, Carnegie Mellon University
Over the last decade, we saw an uprise in the use of machine learning in a wide variety of fields. The availability of large amounts of data coupled with more and more computational power enable researchers and practitioners to train larger and deeper models and thus overcome the limitations of traditional feature extraction approaches. The Human-Computer Interaction (HCI) community is also witnessing this trend. However, as humans have certain expectations of how a system should function, this creates its own unique challenges. Thus, conventional approaches have to be adapted to fit the needs of the HCI community, especially as oftentimes, the input is data directly articulated by humans. In this talk, we will discuss the unique challenges when deploying machine learning in the context of HCI but also how this will ultimately enable us to interact more naturally with everyday devices.
About Sven Mayer
Sven Mayer is a postdoctoral researcher at the Carnegie Mellon University in the Future Interfaces Group. He obtained his PhD degree from the Institute for Visualization and Interactive Systems at the University of Stuttgart in 2019. His research is about modeling of human behavior patterns for interactive systems. Here, he focuses mainly on how to go from simple touch interactions to hand- and body-aware interactions. To accomplish this, he deploys state-of-the-art machine learning techniques to facilitate new interaction techniques.
Machine Learning in Biomedical Imaging – From Cognitive Neuroscience to Computational Psychiatry
Jens Schwarzbach, Universität Regensburg
About Jens Schwarzbach
We investigate cognitive, computational, and metabolic foundations of experience and behavior, especially with respect to psychiatric ailments and disorders, such as depression and anxiety. Complex experiences and behaviors require that brain areas communicate with each other, and that this communication be flexible. By way of functional and structural magnetic resonance imaging, spectroscopy, and also mathematical models, we investigate whether, and in what way, this communication is disturbed in the presence of neuropsychiatric disorders. Furthermore, we investgate how emotions are represented in the brain, and whether a change in the physiological representations leads to altered experience and behavior.