This website is no longer updated.

As of 1.10.2022, the Faculty of Physics has been merged into the TUM School of Natural Sciences with the website For more information read Conversion of Websites.

de | en

Deep Learning in Robotics (IN2349)

Course 0000003810 in SS 2017

General Data

Course Type lecture
Semester Weekly Hours 2 SWS
Organisational Unit Informatics 6 - Chair of Robotics, Artificial Intelligence and Real-time Systems (Prof. Knoll)
Lecturers Berthold Bäuml
Responsible/Coordination: Darius Burschka

Assignment to Modules

Further Information

Courses are together with exams the building blocks for modules. Please keep in mind that information on the contents, learning outcomes and, especially examination conditions are given on the module level only – see section "Assignment to Modules" above.

additional remarks Notice:The lecture is currently only approved for the master's course "Robotics, Cognition, Intelligence” and you might have to check with the student office, if this course will be recognized. The lecture covers the mathematical foundations and the efficient implementation of modern Deep Learning Neural Network Architectures (incl. One-Shot Learning) and its application to problems with real robots (e.g., tactile material classification with a robotic hand). Key points: Overview: Shallow Neural Networks (Perceptrons, Backpropagation, Automatic Differentiation, Autoencoder Networks, Constrained Optimization) Deep Neural Networks (Hierarchical Feature Extraction; Stochastic Optimization, Regularization Methods, Convolutional Neural Networks) Software Frameworks (Google TensorFlow, Facebook Torch, Theano, Caffee, Microsoft CNTK) Applications (“Tactile Material Classification”, “Hand-Eye Coordination for Robotic Grasping”) Robustness (Validation Methods, Networks with Confidence Prediction) Recursive Neural Networks Transfer & One Shot Learning (Siamese Neural Networks, Neural Turing Machines) Excursion to Institute of Robotics and Mechatronics
Links TUMonline entry
Top of page