Organizational Notes

Writing your Bachelor’s or Master’s thesis with us.

Contents

This page clarifies how to start working on your thesis and answers questions that students generally ask (or at least should ask) throughout the process of writing their thesis.

If you plan to write your thesis in our group:

  • You must attend the group seminar.
  • We strongly recommend that you carefully read the study regulations that apply to you beforehand (you might find these here).
  • It is your responsibility to regularly update your supervisor on your progress.

Group Seminar

The group seminar is the joint Bachelor/Master seminar of the Modeling and Simulation group of Saarland University and the DFKI research department Neuro-Mechanistic Modeling. We will invite you to upcoming talks via mail. The current default slot is Tuesday, 11 am.

Law and Order

While you are writing your thesis with us, we assume that you attend all talks. If you have an urgent reason for not attending, you need to excuse yourself in advance. Repeatedly missing talks without an excuse might lead to the termination of your project.

Proposal Paper

Before you can start to work on your thesis, you will write a short proposal paper. It should cover a short introduction into your topic, including its research background. Your paper should:

  • contain a description of your problem,
  • contain all the basic concepts you will need for your thesis,
  • include a discussion of related work,
  • sketch what you plan to work on,
  • especially state what your thesis will cover and what not.

While this might seem annoying at first sight, it actually is quite useful for you: writing the proposal paper will force you to think about what you want to do in your thesis. It will give you some guarantees that we will not extend your thesis topic to infinity, as we have a paper that agrees on what will be done (and what not!). Lastly, it can help us teach you some things about scientific writing before you start writing your thesis, which will spare a lot of work for both you and your advisor.

When your proposal paper is finished, you can schedule your proposal talk.

Proposal Talk

Your proposal talk will be held as soon as your proposal paper was accepted, which means you accomplished the first insights/results of your topic. The talk should last about 25 minutes and include

  • the most important basics of your topic,
  • the most important algorithms/theoretical approaches you plan to use,
  • a short glimpse of your work/your first results.

This proposal talk will be graded as your Bachelor/Master seminar.

Thesis Registration

Shortly after your proposal talk, you have to register your thesis. If you have questions regarding the registration, please address the (correct) examination office. Such things change over time. Thus, if we answer this question, we could give you outdated information, which is the last thing we want to do.

Final Talk / Colloquium Talk

Your final talk will be held as soon as you finished writing your thesis. It should last about 30 minutes and include

  • a short summary about the basics/algorithms/theoretical approaches that you used,
  • a detailed description of what you have done,
  • a detailed presentation of your results.

This final talk will be your Bachelor/Master thesis colloquium.

Further Conditions

Unless otherwise agreed, you must send a first version of your slides to your supervisor one week before your talk at the latest. If you wish to receive detailed feedback, we advise you to do that even earlier.

If you want to schedule your talk or want to be added to the course, please contact us or talk to your supervisor.

Getting Started

  1. Template. We have a template for your thesis, which you can but do not need to use.
  2. Git. If you want to have a repository in our GitLab instance (where you found the template), register there and notify us. Of course, you are also free to use other instances of git, but having the code and thesis at ours makes many things, e.g., feedback, a lot easier.
  3. Meetings & Reports. Not mandatory, but optional and (in some cases highly) recommended, is to agree on a bi-weekly thesis report to your advisor. This report should contain both what you did and what you read (and why the latter helped you/did not help you and what you think about it). For your advisor, this has the benefit of having a better overview of your progress (and therefore the option to intervene, if necessary). For you, this has the benefit of having a detailed literature list when it finally comes to writing your thesis. Further, this is some kind of small external pressure system, which some students tell us is very helpful for them. You should use the thesis template for this report.
  4. Computing Resources. If you need computing resources, there are a few options:
    • If you can use your own machine, this is the easiest way for all. Especially for writing, this is your only option, but this should not require too many resources anyway.
    • If you need to run long-time experiments on CPU, we have a student server we can give you access to. It is a Jupyter server that is accessible through your browser.
    • If you need to run long-time experiments on GPU, you need access to the faculty’s servers managed by Alexander Fauß.

Working With Your Advisor

Your thesis advisor’s role is to advise; not to decide. Sure, there are some decisions you need to make together/agree on. But generally, it is your thesis! You can ask for advice for specific questions or how to continue when you get stuck, but you should not regularly ask your advisor what to do next. A part of scientific working is to figure exactly that out on your own. Rather present your ideas (and in the worst case, your advisor will tell you why they are bad) than regularly ask them what to do.

Writing Your Thesis

When it comes to writing your thesis, there are some aspects you should consider:

  1. Start early. Especially with the basics/preliminaries/methodology chapters, there is no reason to wait until the working phase of your thesis is done.
  2. Gather your advisor’s feedback early. Do not wait until you have written multiple chapters. The advice you receive is much easier to consider if you do not need to rewrite whole chapters.
  3. Apply feedback everywhere. If you receive feedback concerning the scientific writing, apply it to your whole thesis, not just to the chapter you received it for. We are often advising several students and want to spend our time giving you helpful feedback, not marking the same things over and over again.
  4. Present a proper form to your advisor. You want to receive feedback concerning the scientific writing and the content you wrote. This is hard to give if your writing contains a lot of typos or other mistakes.
  5. Do not worry about the alignment and spacings. This is something you do in the very end.

In general, you can expect us to proofread (and give feedback to) your whole thesis once. Reading it several times is just impossible due to time, and reading less is quite unpleasant for you. Of course, if exceptions like rewriting parts of your thesis occur, we also may read that part again, do not worry. So, there are two things you should keep in mind:

  1. You generally receive feedback once! So make sure that the things you give us have a form such that the feedback is valuable for you.
  2. Reading takes time! If you give us your whole thesis shortly before the deadline, it is impossible to read it all. So please give us an appropriate time before you need to hand in your thesis.

See also Typical Mistakes You Should Avoid When Writing Your Thesis.

Reproducibility

One of the most important things in scientific work is the reproducibility of your findings. So make sure to

  • document your code,
  • make it available to your advisors and supervisors, and
  • make it reproducible, e.g., fix random seeds and document them.