Masters Theses and Projects in Machine Learning and Neuromorphic Computing/Learning at the Institute for Theoretical Computer Science
We are looking for a few highly motivated and capable students who want to participate, during their Masters Project and Thesis, in research on areas of Machine Learning and brain-derived paradigms for computing and learning that are currently very active.
In particular, you can become an expert in the new Learning-to-Learn technology, which enables a machine (and also the brain) to learn a new task very fast after having previously learnt a number of related tasks ("transfer learning"). Prof. Legenstein offers a seminar with an introduction to Learning-to-Learn in the WS 2017/18 (708.111).
For Learning-to-Learn experiments, you will learn to use a computer cluster at our Institute. Additionally, in collaboration with partners in the Human Brain Project, you will also learn to use novel brain-inspired computer hardware, as well as a supercomputer in Jülich/Germany. Visits of our master students to the Labs of these research partners are encouraged.
Concrete topics for master projects and theses will be defined in discussion with the student, in order to accomodate prior knowledge and interests. Candidates should have taken the course Computational Intelligence, as well as at least one of the courses among Machine Learning, Neural Networks, Principles of Brain Computation.