Neuronal Deep Learning for Autonomous Systems (Master)

Neuronal Deep Learning for Autonomous Systems introduces Deep Learning from a biological and machine learning perspective. Hence, the course covers the anatomical structure of the brain, the biological learning mechanisms and the respective modelling from the computational neuroscience research. From the machine learning perspective, the course introduces the concepts machine learning briefly and then focuses on the mathematical foundations of linear algebra, optimization- and regularization- techniques used in Deep Learning and the common network architectures. After both perspectives are introduced the applicability to autonomous systems is evaluated. The course finishes with research talks focusing on current state-of-the-art of Deep Learning and the respective applications.

After the lecture “Neuronal Deep Learning for Autonomous Systems”, the students can understand state-of-the-art Deep Learning publications and systems from biology and technology. Furthermore, the students know the constraints of Deep Learning and can apply this knowledge to design the optimal neuron types, architecture and training for a given problem with Deep Learning.