Semester projects for students of the Angewandte Informatik program:
In the summer term 2013 I offer the project "Browser-basierte Datenvisualisierung für maschinelles Lernen".
I am maintaining a list of possible topics for Bachelor and Master theses. Most (but not all) of these projects are related to my lectures on machine learning and evolutionary algorithms. The list of open projects is constantly evolving. Please contact me if you are interested so I can provide you with more details.
I am currently offering a Master's project on modeling of human movement trajectories.
The trajectories have been recorded in the institute's movement laboratory. In this project, the trajectories will be modeled as superpositions of relatively few basic components. Both the compoents and the trajectory-wise coefficients are to be learned from data, which can be achieved with variants of algorithms like independent component analysis (ICA) and learning methods for sparse reconstruction.
The goal is to uncover movements components in a purely data-driver manner that allow for an interpretation as meaningful "movement primitives". The explanatory value of this model will be assessed by learning theoretical and statistical measures, and also by comparing the results to psycho-physical models.
One line of my research deals with supervised learning with support vector machines (SVMs). On the one hand I am interested in the SVM training problem, which basically amounts to large scale quadratic programming. On the other hand I am trying to simplify SVM usage for non-experts by developing robust methods for automatic model selection. My research activities include both theoretical and practical aspects ranging from SVM optimization to experimental comparison studies and software development.
I am also interested in evolutionary algorithms, in particular evolution strategies for controller design and real-valued black box optimization. This activity offers fascinating theoretical challenges, and at the same time highly efficient practical algorithms with endless applications.
I am an active developer of the Shark Machine Learning Library. Shark is an open-source, modular, and fast C++ library. A large share of my research code is either part of the library or based thereon. Check it out!
Shark is currently undergoing a major transition; actually it is more fair to speak of a complete rewrite. By now (2013) the work is mostly done, and we already have an alpha release of the brand new Shark 3. A few more design changes are underway that will make the library even faster and even more complete.