masters-thesis

preface.tex (2883B)

---

 \chapter*{Preface}
 \addcontentsline{toc}{chapter}{Preface} % Add the preface to the table of contents as a chapter
 
 I've been studying for a pretty long time.
 Because of this, I suppose there is some sense of anticipation, that your thesis is the culmination of all the work you've done up to this point.
 Having not always been the biggest adherent to the idea of symbol-for-the-sake-of-symbol, I wasn't sure what to expect of this experience.
 For many, writing a thesis seems to be a mixed bag, though in general people feel a sense of accomplishment, since they see it as being the last step in a long journey.
 
 I definitely had a slow start to the thesis (most people do), since for the first month or so I just explored the software and the literature to get a feel for the problem.
 Looking back, this was a really good idea, since understanding the software ended up being key to solving the problems in this thesis.
 When I tried to generate some preliminary results, I quickly encountered some roadblocks.
 As it turned out, though, these issues were quite fundamental to the problem---but my supervisor, S{\o}ren, had found a paper which addressed some of them.
 He then invited me to a hackathon with his research group, where they were planning to sit together and try to implement the methods from the paper.
 Initially, I was a bit hesitant about going, since I had never been to a research-related hackathon, and I was a bit worried that I would be out of my depth and that I wouldn't be able to contribute.
 However, after Lorenza helped me get over my fears, I decided to go, and I'm glad I did.
 
 The hackathon and the project that followed it were probably the most fun I've had while studying at DTU.
 I got the chance to work on some really novel problems (like dealing with ill conditioning and numerical issues) and enjoyed every minute of it.
 
 This thesis is the product of a five-month-long span of work conducted in the Section for Cognitive Systems of DTU Compute, during which I helped develop some of the ideas in the ``Large-Scale Laplace'' project.
 This project is, at the time of writing, still ongoing; we'll hopefully soon publish the results of our work.
 In this project, I developed and implemented the sampling procedure for the full Laplace approximation.
 This includes the implementation of the sampling algorithm (and all the mathematical machinery behind it), as well as the formulation, implementation and testing of preconditioners.
 During the hackathon, I collaborated with Marco Miani and Eugene Zainchkovskyy on the early work on sampling.
 I also contributed to parts of the derivation and implementation of the marginal training algorithm, though I would say the bulk of this was done by Hrittik Roy, Johan Ye and Pablo Moreno-Mu\~noz.
 And S{\o}ren Hauberg helped massively everywhere.
 
 \begin{flushright}
 	\textit{Silas Brack}
 \end{flushright}