Perceiving Neural Networks
Perception is a computational feat. The conversion of high-dimensional sensory input to meaning relies on the ability to solve complex pattern recognition problems. Natural tasks like object recognition or visual search are good examples of this process revealing the computational challenges underlying perception. To tackle these challenges complex neural networks have developed in the brain that perform surprisingly well. At the interface between artificial intelligence and neuroscience we focus on uncovering the algorithms and neuro-computational design principles of perceiving neural networks. A practical example of the outcome of this research is a new method for creating artistic images. More specifically, we want to explain how characteristic properties of neural systems originate from the computational requirements of specific perceptual skills.
Machine Learning



Academic Partnerships in Tübingen
External Academic Partnerships
Spin-offs
- Layer7 AI (founded by Matthias Bethge, Alexander Ecker, Wieland Brendel, Peter Droege, Ingo Hoffmann, Behar Veliqi & Isabel Suditsch)
- DeepArt (co-founded by Matthias Bethge, Alexander Ecker & Leon Gatys)
Industry Partnerships
- Twitter Cortex London (former Magic Pony Technology)
- Grammofy (co-founded by Matthias Kümmerer)
- Facebook AI Research (GPU Partnership)
- Intel Network on Intelligent Systems
- Adobe Research (Fellowship award to Leon Gatys)
- AmbiGate (EXIST mentor)
- Paperspace (Cloud GPUs)