An important area within the field of computer science is the study of algorithms. Algorithms describe programmatic ways to solve a problem or a class of problems. But there are always differences between the theoretical and the practical solutions. Such differences may be more critical or less critical depending on the application. This thesis explores how adding robustness to known problems affects algorithmic complexity. It focuses on the so-called ε-robust problems which are problems where the required robustness is parameterized by some constant ε. It introduces a theoretical framework and can be applied to almost any robust version of any problem, as long as the robustness (in terms of ε) is defined according to a couple of rules. This can find application in a number of real-life scenarios, ranging from Computer-Aided Design and Sensor Networks, to Robotics and Motion Planning. The thesis is well organized, with all the concepts and theoretical findings neatly defined, leaving no gaps to the reader. It forms the basis of a research paper that is to be sent to STOC or FLOC, that are top-tier conferences in theoretical computer science. Overall, it is an exceptional piece of work that is distinguished by its theoretical rigor.
After scoring all the thesis on 4 different categories (scientific quality, technical quality, societal impact and scientific impact), Lammert his thesis scored the highest on average although the final scores were really close. Lammert his thesis stood out from the others because he took an underlighted subject that should receive more (scientific) attention and wrote an outstanding thesis about it.
