Every PhD student of the Research Training Group 1023 is associated to an individual team of supervisors that consists of three professors including the supervisor of the PhD project. At the beginning of every academic year, every PhD student, consulting the team of supervisors, develops an individual annual plan stating the courses he wants to visit and the research objectives he plans to achieve. At the end of the academic year, the PhD student writes a progress report informing the team of supervisors about his progress in research and studies. The PhD students can always contact their supervisors in case of technical or formal problems.
The study programm consists of three one-year periods:
During this period the PhD students should attend some canonical courses in applied mathematics in order to obtain a broad basic knowledge in the field of stochastics as well as in numerical mathematics. Courses in functional analysis, optimization, measure and probability theory, and statistical data analysis are offered every year, consisting of 4 hours of lectures and exercises. A course in partial differential equations, consisting of 4 hours of lectures and exercises, is offered at least every two years. The individual advice of the team of supervisors assures that in the second year the PhD student can participate in more advanced courses. Depending on the state of knowledge of the student, in the first year the program of studies should include up to 8 hours per week.
At least 4 hours per week, courses in fields like statistical inverse problems, applied stochastic processes, image processing, algorithmic and statistical learning, convex optimization, or multi-scale methods are offered.
These courses consist of 2 or 4 hours of lectures, in some cases compact courses are offered. In this period, the program of studies should consist of 4-6 hours per semester.
In the last year of the PhD program in accordance with the supervisors the students should attend 4-6 hours per week of special lectures and seminars in the area of their PhD thesis. On behalf of a fast completion of the PhD studies, the courses should be strongly related to the subject of the thesis. That could for example be courses in fields like convex optimization in regularization, statistical analysis of high-dimensional problems, pattern recognition in biometry (Area A), probabilistic data models, compression methods (Area B), or inverse scattering theory and optimization with partial differential equations (Area C).